@ssml-builder-js/azure-tts-client 2.4.0 → 2.6.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (14) hide show
  1. package/CHANGELOG.md +12 -0
  2. package/dist/index.d.mts +4 -0
  3. package/dist/index.d.ts +4 -0
  4. package/dist/index.js +28 -2
  5. package/dist/index.js.map +1 -1
  6. package/dist/index.mjs +28 -2
  7. package/dist/index.mjs.map +1 -1
  8. package/package.json +1 -1
  9. package/src/client.ts +2 -2
  10. package/src/synthesis.ts +27 -0
  11. package/src/types.ts +4 -0
  12. package/test/edge-cases.test.ts +43 -0
  13. package/test/output-formats.test.ts +8 -0
  14. package/test/synthesis.test.ts +100 -0
package/CHANGELOG.md CHANGED
@@ -1,5 +1,17 @@
1
1
  # @ssml-builder-js/azure-tts-client
2
2
 
3
+ ## 2.6.0
4
+
5
+ ### Patch Changes
6
+
7
+ - Add synthesis cancellation, timeout, and optionally enabled live Azure Speech API coverage tests.
8
+
9
+ ## 2.5.0
10
+
11
+ ### Minor Changes
12
+
13
+ - Add timeout and cancellation support for synthesis requests and document low-bitrate MP3 and server-side integration patterns.
14
+
3
15
  ## 2.4.0
4
16
 
5
17
  ### Minor Changes
package/dist/index.d.mts CHANGED
@@ -1,4 +1,6 @@
1
1
  interface TtsConfig {
2
+ signal?: AbortSignal;
3
+ timeoutMs?: number;
2
4
  endpoint?: string;
3
5
  subscriptionKey: string;
4
6
  region: string;
@@ -11,6 +13,8 @@ interface AzureTtsLogger {
11
13
  error?: (...args: unknown[]) => void;
12
14
  }
13
15
  interface AzureTtsClientOptions {
16
+ signal?: AbortSignal;
17
+ timeoutMs?: number;
14
18
  subscriptionKey: string;
15
19
  region: string;
16
20
  endpoint?: string;
package/dist/index.d.ts CHANGED
@@ -1,4 +1,6 @@
1
1
  interface TtsConfig {
2
+ signal?: AbortSignal;
3
+ timeoutMs?: number;
2
4
  endpoint?: string;
3
5
  subscriptionKey: string;
4
6
  region: string;
@@ -11,6 +13,8 @@ interface AzureTtsLogger {
11
13
  error?: (...args: unknown[]) => void;
12
14
  }
13
15
  interface AzureTtsClientOptions {
16
+ signal?: AbortSignal;
17
+ timeoutMs?: number;
14
18
  subscriptionKey: string;
15
19
  region: string;
16
20
  endpoint?: string;
package/dist/index.js CHANGED
@@ -151,30 +151,56 @@ function closeSpeechResources(speechConfig, synthesizer) {
151
151
  }
152
152
  }
153
153
  async function synthesizeSpeech(ssml, config) {
154
+ if (config.signal?.aborted) {
155
+ throw createSpeechSdkError("Speech synthesis was cancelled.");
156
+ }
154
157
  const speechConfig = createSpeechConfig(config);
155
158
  const synthesizer = new SpeechSDK2.SpeechSynthesizer(speechConfig, null);
156
159
  return await new Promise((resolve, reject) => {
157
160
  let resourcesClosed = false;
161
+ let settled = false;
162
+ let timeout;
163
+ let abortHandler;
164
+ const cleanup = () => {
165
+ if (timeout) clearTimeout(timeout);
166
+ if (abortHandler) config.signal?.removeEventListener("abort", abortHandler);
167
+ };
158
168
  const closeResources = () => {
159
169
  if (resourcesClosed) return;
160
170
  resourcesClosed = true;
161
171
  closeSpeechResources(speechConfig, synthesizer);
162
172
  };
163
173
  const rejectWithError = (error) => {
174
+ if (settled) return;
175
+ settled = true;
176
+ cleanup();
164
177
  closeResources();
165
178
  reject(createSpeechSdkError(error));
166
179
  };
167
180
  const cb = (result) => {
181
+ if (settled) return;
168
182
  const { reason, errorDetails } = result;
169
183
  if (reason !== SpeechSDK2.ResultReason.SynthesizingAudioCompleted) {
170
184
  const err = errorDetails || `Speech synthesis failed with reason ${reason}.`;
171
185
  rejectWithError(err);
172
186
  return;
173
187
  }
188
+ settled = true;
189
+ cleanup();
174
190
  closeResources();
175
191
  resolve(result.audioData);
176
192
  };
177
193
  try {
194
+ if (config.signal) {
195
+ abortHandler = () => rejectWithError("Speech synthesis was cancelled.");
196
+ config.signal.addEventListener("abort", abortHandler, { once: true });
197
+ }
198
+ if (config.timeoutMs !== void 0 && config.timeoutMs > 0) {
199
+ timeout = setTimeout(
200
+ () => rejectWithError(`Speech synthesis timed out after ${config.timeoutMs} ms.`),
201
+ config.timeoutMs
202
+ );
203
+ }
178
204
  synthesizer.speakSsmlAsync(ssml, cb, rejectWithError);
179
205
  } catch (error) {
180
206
  rejectWithError(error);
@@ -191,10 +217,10 @@ var AzureTtsClient = class {
191
217
  __privateSet(this, _options, options);
192
218
  }
193
219
  async synthesize(ssml) {
194
- const { region, subscriptionKey, outputFormat } = __privateGet(this, _options);
220
+ const { region, subscriptionKey, outputFormat, signal, timeoutMs } = __privateGet(this, _options);
195
221
  const endpoint = __privateGet(this, _options).endpoint?.trim() || ENDPOINT_TEMPLATE.replace("{region}", region);
196
222
  __privateGet(this, _options).logger?.debug?.("Using Azure TTS endpoint:", endpoint);
197
- const config = { endpoint, region, subscriptionKey, outputFormat };
223
+ const config = { endpoint, region, subscriptionKey, outputFormat, signal, timeoutMs };
198
224
  return synthesizeSpeech(ssml, config);
199
225
  }
200
226
  };
package/dist/index.js.map CHANGED
@@ -1 +1 @@
1
- {"version":3,"sources":["../src/index.ts","../src/errors.ts","../src/synthesis.ts","../src/speechConfig.ts","../src/outputFormats.ts","../src/client.ts"],"sourcesContent":["/**\n * azure-tts-client: Azure Text-to-Speech client for SSML playback.\n */\n\nexport type { AzureTtsClientOptions, AzureTtsLogger, TtsConfig } from \"./types.ts\";\nexport { AzureTtsError, AzureTtsSdkError } from \"./errors.ts\";\nexport { AzureTtsClient } from \"./client.ts\";\nexport { synthesizeSpeech } from \"./synthesis.ts\";\n","export class AzureTtsError extends Error {\n readonly status: number;\n readonly statusText: string;\n readonly responseBody: string;\n readonly requestId: string | null;\n\n constructor(status: number, statusText: string, responseBody: string, requestId: string | null) {\n super(`Azure TTS request failed: ${status} ${statusText}`);\n this.name = \"AzureTtsError\";\n this.status = status;\n this.statusText = statusText;\n this.responseBody = responseBody;\n this.requestId = requestId;\n }\n}\n\nexport class AzureTtsSdkError extends AzureTtsError {\n readonly errorDetails: string;\n\n constructor(errorDetails: string) {\n super(0, \"Speech SDK\", errorDetails, null);\n this.name = \"AzureTtsSdkError\";\n this.message = `Azure TTS synthesis failed: ${errorDetails}`;\n this.errorDetails = errorDetails;\n }\n}\n\nexport function createSpeechSdkError(error: unknown): AzureTtsSdkError {\n const message = error instanceof Error ? error.message : String(error);\n return new AzureTtsSdkError(message);\n}\n","import * as SpeechSDK from \"microsoft-cognitiveservices-speech-sdk\";\nimport { createSpeechSdkError } from \"./errors.ts\";\nimport { createSpeechConfig } from \"./speechConfig.ts\";\nimport type { TtsConfig } from \"./types.ts\";\n\nfunction closeSpeechResources(speechConfig: SpeechSDK.SpeechConfig, synthesizer: SpeechSDK.SpeechSynthesizer): void {\n try {\n synthesizer.close();\n } catch {}\n\n try {\n speechConfig.close();\n } catch {}\n}\n\nexport async function synthesizeSpeech(ssml: string, config: TtsConfig): Promise<ArrayBuffer> {\n const speechConfig = createSpeechConfig(config);\n const synthesizer = new SpeechSDK.SpeechSynthesizer(speechConfig, null);\n\n return await new Promise<ArrayBuffer>((resolve, reject) => {\n let resourcesClosed = false;\n const closeResources = () => {\n if (resourcesClosed) return;\n resourcesClosed = true;\n closeSpeechResources(speechConfig, synthesizer);\n };\n const rejectWithError = (error: unknown) => {\n closeResources();\n reject(createSpeechSdkError(error));\n };\n\n const cb = (result: SpeechSDK.SpeechSynthesisResult) => {\n const { reason, errorDetails } = result;\n if (reason !== SpeechSDK.ResultReason.SynthesizingAudioCompleted) {\n const err = errorDetails || `Speech synthesis failed with reason ${reason}.`;\n rejectWithError(err);\n return;\n }\n closeResources();\n resolve(result.audioData);\n };\n\n try {\n synthesizer.speakSsmlAsync(ssml, cb, rejectWithError);\n } catch (error) {\n rejectWithError(error);\n }\n });\n}\n","import { SpeechConfig } from \"microsoft-cognitiveservices-speech-sdk\";\nimport { DEFAULT_OUTPUT_FORMAT, resolveOutputFormat } from \"./outputFormats.ts\";\nimport type { TtsConfig } from \"./types.ts\";\n\nexport function resolveEndpoint(config: TtsConfig): string {\n const endpoint = config.endpoint?.trim() || \"https://{region}.tts.speech.microsoft.com/cognitiveservices/v1\";\n return endpoint.replace(/\\{region\\}/g, encodeURIComponent(config.region));\n}\n\nexport function createSpeechConfig(config: TtsConfig): SpeechConfig {\n const { outputFormat = DEFAULT_OUTPUT_FORMAT, subscriptionKey } = config;\n\n const endpoint = new URL(resolveEndpoint(config));\n const speechConfig = SpeechConfig.fromEndpoint(endpoint, subscriptionKey);\n speechConfig.speechSynthesisOutputFormat = resolveOutputFormat(outputFormat);\n return speechConfig;\n}\n","import * as SpeechSDK from \"microsoft-cognitiveservices-speech-sdk\";\n\nexport const DEFAULT_OUTPUT_FORMAT = \"audio-16khz-128kbitrate-mono-mp3\";\n\nconst OUTPUT_FORMATS: Record<string, SpeechSDK.SpeechSynthesisOutputFormat> = {\n \"raw-8khz-8bit-mono-mulaw\": SpeechSDK.SpeechSynthesisOutputFormat.Raw8Khz8BitMonoMULaw,\n \"riff-16khz-16kbps-mono-siren\": SpeechSDK.SpeechSynthesisOutputFormat.Riff16Khz16KbpsMonoSiren,\n \"audio-16khz-16kbps-mono-siren\": SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz16KbpsMonoSiren,\n \"audio-16khz-32kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz32KBitRateMonoMp3,\n \"audio-16khz-128kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz128KBitRateMonoMp3,\n \"audio-16khz-64kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz64KBitRateMonoMp3,\n \"audio-24khz-48kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz48KBitRateMonoMp3,\n \"audio-24khz-96kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz96KBitRateMonoMp3,\n \"audio-24khz-160kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz160KBitRateMonoMp3,\n \"raw-16khz-16bit-mono-truesilk\": SpeechSDK.SpeechSynthesisOutputFormat.Raw16Khz16BitMonoTrueSilk,\n \"riff-16khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff16Khz16BitMonoPcm,\n \"riff-8khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff8Khz16BitMonoPcm,\n \"riff-24khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff24Khz16BitMonoPcm,\n \"riff-8khz-8bit-mono-mulaw\": SpeechSDK.SpeechSynthesisOutputFormat.Riff8Khz8BitMonoMULaw,\n \"raw-16khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw16Khz16BitMonoPcm,\n \"raw-24khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw24Khz16BitMonoPcm,\n \"raw-8khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw8Khz16BitMonoPcm,\n \"ogg-16khz-16bit-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Ogg16Khz16BitMonoOpus,\n \"ogg-24khz-16bit-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Ogg24Khz16BitMonoOpus,\n \"raw-48khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw48Khz16BitMonoPcm,\n \"riff-48khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff48Khz16BitMonoPcm,\n \"audio-48khz-96kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio48Khz96KBitRateMonoMp3,\n \"audio-48khz-192kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio48Khz192KBitRateMonoMp3,\n \"ogg-48khz-16bit-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Ogg48Khz16BitMonoOpus,\n \"webm-16khz-16bit-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Webm16Khz16BitMonoOpus,\n \"webm-24khz-16bit-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Webm24Khz16BitMonoOpus,\n \"webm-24khz-16bit-24kbps-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Webm24Khz16Bit24KbpsMonoOpus,\n \"raw-24khz-16bit-mono-truesilk\": SpeechSDK.SpeechSynthesisOutputFormat.Raw24Khz16BitMonoTrueSilk,\n \"raw-8khz-8bit-mono-alaw\": SpeechSDK.SpeechSynthesisOutputFormat.Raw8Khz8BitMonoALaw,\n \"riff-8khz-8bit-mono-alaw\": SpeechSDK.SpeechSynthesisOutputFormat.Riff8Khz8BitMonoALaw,\n \"audio-16khz-16bit-32kbps-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz16Bit32KbpsMonoOpus,\n \"audio-24khz-16bit-48kbps-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz16Bit48KbpsMonoOpus,\n \"audio-24khz-16bit-24kbps-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz16Bit24KbpsMonoOpus,\n \"raw-22050hz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw22050Hz16BitMonoPcm,\n \"riff-22050hz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff22050Hz16BitMonoPcm,\n \"raw-44100hz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw44100Hz16BitMonoPcm,\n \"riff-44100hz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff44100Hz16BitMonoPcm,\n \"amr-wb-16000hz\": SpeechSDK.SpeechSynthesisOutputFormat.AmrWb16000Hz,\n \"g722-16khz-64kbps\": SpeechSDK.SpeechSynthesisOutputFormat.G72216Khz64Kbps,\n};\n\nexport function resolveOutputFormat(outputFormat: string): SpeechSDK.SpeechSynthesisOutputFormat {\n const resolvedFormat = OUTPUT_FORMATS[outputFormat];\n if (resolvedFormat === undefined) {\n throw new Error(`Unsupported Azure Speech output format: ${outputFormat}`);\n }\n\n return resolvedFormat;\n}\n","import { synthesizeSpeech } from \"./synthesis.ts\";\nimport type { AzureTtsClientOptions } from \"./types.ts\";\n\nconst ENDPOINT_TEMPLATE = \"https://{region}.tts.speech.microsoft.com/cognitiveservices/v1\";\n\nexport class AzureTtsClient {\n readonly #options: AzureTtsClientOptions;\n\n constructor(options: AzureTtsClientOptions) {\n this.#options = options;\n }\n\n async synthesize(ssml: string): Promise<ArrayBuffer> {\n const { region, subscriptionKey, outputFormat } = this.#options;\n const endpoint = this.#options.endpoint?.trim() || ENDPOINT_TEMPLATE.replace(\"{region}\", region);\n this.#options.logger?.debug?.(\"Using Azure TTS endpoint:\", endpoint);\n\n const config = { endpoint, region, subscriptionKey, outputFormat };\n return synthesizeSpeech(ssml, config);\n }\n}\n"],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;;;ACAO,IAAM,gBAAN,cAA4B,MAAM;AAAA,EAMvC,YAAY,QAAgB,YAAoB,cAAsB,WAA0B;AAC9F,UAAM,6BAA6B,MAAM,IAAI,UAAU,EAAE;AACzD,SAAK,OAAO;AACZ,SAAK,SAAS;AACd,SAAK,aAAa;AAClB,SAAK,eAAe;AACpB,SAAK,YAAY;AAAA,EACnB;AACF;AAEO,IAAM,mBAAN,cAA+B,cAAc;AAAA,EAGlD,YAAY,cAAsB;AAChC,UAAM,GAAG,cAAc,cAAc,IAAI;AACzC,SAAK,OAAO;AACZ,SAAK,UAAU,+BAA+B,YAAY;AAC1D,SAAK,eAAe;AAAA,EACtB;AACF;AAEO,SAAS,qBAAqB,OAAkC;AACrE,QAAM,UAAU,iBAAiB,QAAQ,MAAM,UAAU,OAAO,KAAK;AACrE,SAAO,IAAI,iBAAiB,OAAO;AACrC;;;AC9BA,IAAAA,aAA2B;;;ACA3B,oDAA6B;;;ACA7B,gBAA2B;AAEpB,IAAM,wBAAwB;AAErC,IAAM,iBAAwE;AAAA,EAC5E,4BAAsC,sCAA4B;AAAA,EAClE,gCAA0C,sCAA4B;AAAA,EACtE,iCAA2C,sCAA4B;AAAA,EACvE,mCAA6C,sCAA4B;AAAA,EACzE,oCAA8C,sCAA4B;AAAA,EAC1E,mCAA6C,sCAA4B;AAAA,EACzE,mCAA6C,sCAA4B;AAAA,EACzE,mCAA6C,sCAA4B;AAAA,EACzE,oCAA8C,sCAA4B;AAAA,EAC1E,iCAA2C,sCAA4B;AAAA,EACvE,6BAAuC,sCAA4B;AAAA,EACnE,4BAAsC,sCAA4B;AAAA,EAClE,6BAAuC,sCAA4B;AAAA,EACnE,6BAAuC,sCAA4B;AAAA,EACnE,4BAAsC,sCAA4B;AAAA,EAClE,4BAAsC,sCAA4B;AAAA,EAClE,2BAAqC,sCAA4B;AAAA,EACjE,6BAAuC,sCAA4B;AAAA,EACnE,6BAAuC,sCAA4B;AAAA,EACnE,4BAAsC,sCAA4B;AAAA,EAClE,6BAAuC,sCAA4B;AAAA,EACnE,mCAA6C,sCAA4B;AAAA,EACzE,oCAA8C,sCAA4B;AAAA,EAC1E,6BAAuC,sCAA4B;AAAA,EACnE,8BAAwC,sCAA4B;AAAA,EACpE,8BAAwC,sCAA4B;AAAA,EACpE,qCAA+C,sCAA4B;AAAA,EAC3E,iCAA2C,sCAA4B;AAAA,EACvE,2BAAqC,sCAA4B;AAAA,EACjE,4BAAsC,sCAA4B;AAAA,EAClE,sCAAgD,sCAA4B;AAAA,EAC5E,sCAAgD,sCAA4B;AAAA,EAC5E,sCAAgD,sCAA4B;AAAA,EAC5E,8BAAwC,sCAA4B;AAAA,EACpE,+BAAyC,sCAA4B;AAAA,EACrE,8BAAwC,sCAA4B;AAAA,EACpE,+BAAyC,sCAA4B;AAAA,EACrE,kBAA4B,sCAA4B;AAAA,EACxD,qBAA+B,sCAA4B;AAC7D;AAEO,SAAS,oBAAoB,cAA6D;AAC/F,QAAM,iBAAiB,eAAe,YAAY;AAClD,MAAI,mBAAmB,QAAW;AAChC,UAAM,IAAI,MAAM,2CAA2C,YAAY,EAAE;AAAA,EAC3E;AAEA,SAAO;AACT;;;ADjDO,SAAS,gBAAgB,QAA2B;AACzD,QAAM,WAAW,OAAO,UAAU,KAAK,KAAK;AAC5C,SAAO,SAAS,QAAQ,eAAe,mBAAmB,OAAO,MAAM,CAAC;AAC1E;AAEO,SAAS,mBAAmB,QAAiC;AAClE,QAAM,EAAE,eAAe,uBAAuB,gBAAgB,IAAI;AAElE,QAAM,WAAW,IAAI,IAAI,gBAAgB,MAAM,CAAC;AAChD,QAAM,eAAe,2DAAa,aAAa,UAAU,eAAe;AACxE,eAAa,8BAA8B,oBAAoB,YAAY;AAC3E,SAAO;AACT;;;ADXA,SAAS,qBAAqB,cAAsC,aAAgD;AAClH,MAAI;AACF,gBAAY,MAAM;AAAA,EACpB,QAAQ;AAAA,EAAC;AAET,MAAI;AACF,iBAAa,MAAM;AAAA,EACrB,QAAQ;AAAA,EAAC;AACX;AAEA,eAAsB,iBAAiB,MAAc,QAAyC;AAC5F,QAAM,eAAe,mBAAmB,MAAM;AAC9C,QAAM,cAAc,IAAc,6BAAkB,cAAc,IAAI;AAEtE,SAAO,MAAM,IAAI,QAAqB,CAAC,SAAS,WAAW;AACzD,QAAI,kBAAkB;AACtB,UAAM,iBAAiB,MAAM;AAC3B,UAAI,gBAAiB;AACrB,wBAAkB;AAClB,2BAAqB,cAAc,WAAW;AAAA,IAChD;AACA,UAAM,kBAAkB,CAAC,UAAmB;AAC1C,qBAAe;AACf,aAAO,qBAAqB,KAAK,CAAC;AAAA,IACpC;AAEA,UAAM,KAAK,CAAC,WAA4C;AACtD,YAAM,EAAE,QAAQ,aAAa,IAAI;AACjC,UAAI,WAAqB,wBAAa,4BAA4B;AAChE,cAAM,MAAM,gBAAgB,uCAAuC,MAAM;AACzE,wBAAgB,GAAG;AACnB;AAAA,MACF;AACA,qBAAe;AACf,cAAQ,OAAO,SAAS;AAAA,IAC1B;AAEA,QAAI;AACF,kBAAY,eAAe,MAAM,IAAI,eAAe;AAAA,IACtD,SAAS,OAAO;AACd,sBAAgB,KAAK;AAAA,IACvB;AAAA,EACF,CAAC;AACH;;;AG7CA,IAAM,oBAAoB;AAH1B;AAKO,IAAM,iBAAN,MAAqB;AAAA,EAG1B,YAAY,SAAgC;AAF5C,uBAAS;AAGP,uBAAK,UAAW;AAAA,EAClB;AAAA,EAEA,MAAM,WAAW,MAAoC;AACnD,UAAM,EAAE,QAAQ,iBAAiB,aAAa,IAAI,mBAAK;AACvD,UAAM,WAAW,mBAAK,UAAS,UAAU,KAAK,KAAK,kBAAkB,QAAQ,YAAY,MAAM;AAC/F,uBAAK,UAAS,QAAQ,QAAQ,6BAA6B,QAAQ;AAEnE,UAAM,SAAS,EAAE,UAAU,QAAQ,iBAAiB,aAAa;AACjE,WAAO,iBAAiB,MAAM,MAAM;AAAA,EACtC;AACF;AAdW;","names":["SpeechSDK"]}
1
+ {"version":3,"sources":["../src/index.ts","../src/errors.ts","../src/synthesis.ts","../src/speechConfig.ts","../src/outputFormats.ts","../src/client.ts"],"sourcesContent":["/**\n * azure-tts-client: Azure Text-to-Speech client for SSML playback.\n */\n\nexport type { AzureTtsClientOptions, AzureTtsLogger, TtsConfig } from \"./types.ts\";\nexport { AzureTtsError, AzureTtsSdkError } from \"./errors.ts\";\nexport { AzureTtsClient } from \"./client.ts\";\nexport { synthesizeSpeech } from \"./synthesis.ts\";\n","export class AzureTtsError extends Error {\n readonly status: number;\n readonly statusText: string;\n readonly responseBody: string;\n readonly requestId: string | null;\n\n constructor(status: number, statusText: string, responseBody: string, requestId: string | null) {\n super(`Azure TTS request failed: ${status} ${statusText}`);\n this.name = \"AzureTtsError\";\n this.status = status;\n this.statusText = statusText;\n this.responseBody = responseBody;\n this.requestId = requestId;\n }\n}\n\nexport class AzureTtsSdkError extends AzureTtsError {\n readonly errorDetails: string;\n\n constructor(errorDetails: string) {\n super(0, \"Speech SDK\", errorDetails, null);\n this.name = \"AzureTtsSdkError\";\n this.message = `Azure TTS synthesis failed: ${errorDetails}`;\n this.errorDetails = errorDetails;\n }\n}\n\nexport function createSpeechSdkError(error: unknown): AzureTtsSdkError {\n const message = error instanceof Error ? error.message : String(error);\n return new AzureTtsSdkError(message);\n}\n","import * as SpeechSDK from \"microsoft-cognitiveservices-speech-sdk\";\nimport { createSpeechSdkError } from \"./errors.ts\";\nimport { createSpeechConfig } from \"./speechConfig.ts\";\nimport type { TtsConfig } from \"./types.ts\";\n\nfunction closeSpeechResources(speechConfig: SpeechSDK.SpeechConfig, synthesizer: SpeechSDK.SpeechSynthesizer): void {\n try {\n synthesizer.close();\n } catch {}\n\n try {\n speechConfig.close();\n } catch {}\n}\n\nexport async function synthesizeSpeech(ssml: string, config: TtsConfig): Promise<ArrayBuffer> {\n if (config.signal?.aborted) {\n throw createSpeechSdkError(\"Speech synthesis was cancelled.\");\n }\n\n const speechConfig = createSpeechConfig(config);\n const synthesizer = new SpeechSDK.SpeechSynthesizer(speechConfig, null);\n\n return await new Promise<ArrayBuffer>((resolve, reject) => {\n let resourcesClosed = false;\n let settled = false;\n let timeout: ReturnType<typeof setTimeout> | undefined;\n let abortHandler: (() => void) | undefined;\n const cleanup = () => {\n if (timeout) clearTimeout(timeout);\n if (abortHandler) config.signal?.removeEventListener(\"abort\", abortHandler);\n };\n const closeResources = () => {\n if (resourcesClosed) return;\n resourcesClosed = true;\n closeSpeechResources(speechConfig, synthesizer);\n };\n const rejectWithError = (error: unknown) => {\n if (settled) return;\n settled = true;\n cleanup();\n closeResources();\n reject(createSpeechSdkError(error));\n };\n\n const cb = (result: SpeechSDK.SpeechSynthesisResult) => {\n if (settled) return;\n const { reason, errorDetails } = result;\n if (reason !== SpeechSDK.ResultReason.SynthesizingAudioCompleted) {\n const err = errorDetails || `Speech synthesis failed with reason ${reason}.`;\n rejectWithError(err);\n return;\n }\n settled = true;\n cleanup();\n closeResources();\n resolve(result.audioData);\n };\n\n try {\n if (config.signal) {\n abortHandler = () => rejectWithError(\"Speech synthesis was cancelled.\");\n config.signal.addEventListener(\"abort\", abortHandler, { once: true });\n }\n if (config.timeoutMs !== undefined && config.timeoutMs > 0) {\n timeout = setTimeout(\n () => rejectWithError(`Speech synthesis timed out after ${config.timeoutMs} ms.`),\n config.timeoutMs,\n );\n }\n synthesizer.speakSsmlAsync(ssml, cb, rejectWithError);\n } catch (error) {\n rejectWithError(error);\n }\n });\n}\n","import { SpeechConfig } from \"microsoft-cognitiveservices-speech-sdk\";\nimport { DEFAULT_OUTPUT_FORMAT, resolveOutputFormat } from \"./outputFormats.ts\";\nimport type { TtsConfig } from \"./types.ts\";\n\nexport function resolveEndpoint(config: TtsConfig): string {\n const endpoint = config.endpoint?.trim() || \"https://{region}.tts.speech.microsoft.com/cognitiveservices/v1\";\n return endpoint.replace(/\\{region\\}/g, encodeURIComponent(config.region));\n}\n\nexport function createSpeechConfig(config: TtsConfig): SpeechConfig {\n const { outputFormat = DEFAULT_OUTPUT_FORMAT, subscriptionKey } = config;\n\n const endpoint = new URL(resolveEndpoint(config));\n const speechConfig = SpeechConfig.fromEndpoint(endpoint, subscriptionKey);\n speechConfig.speechSynthesisOutputFormat = resolveOutputFormat(outputFormat);\n return speechConfig;\n}\n","import * as SpeechSDK from \"microsoft-cognitiveservices-speech-sdk\";\n\nexport const DEFAULT_OUTPUT_FORMAT = \"audio-16khz-128kbitrate-mono-mp3\";\n\nconst OUTPUT_FORMATS: Record<string, SpeechSDK.SpeechSynthesisOutputFormat> = {\n \"raw-8khz-8bit-mono-mulaw\": SpeechSDK.SpeechSynthesisOutputFormat.Raw8Khz8BitMonoMULaw,\n \"riff-16khz-16kbps-mono-siren\": SpeechSDK.SpeechSynthesisOutputFormat.Riff16Khz16KbpsMonoSiren,\n \"audio-16khz-16kbps-mono-siren\": SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz16KbpsMonoSiren,\n \"audio-16khz-32kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz32KBitRateMonoMp3,\n \"audio-16khz-128kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz128KBitRateMonoMp3,\n \"audio-16khz-64kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz64KBitRateMonoMp3,\n \"audio-24khz-48kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz48KBitRateMonoMp3,\n \"audio-24khz-96kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz96KBitRateMonoMp3,\n \"audio-24khz-160kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz160KBitRateMonoMp3,\n \"raw-16khz-16bit-mono-truesilk\": SpeechSDK.SpeechSynthesisOutputFormat.Raw16Khz16BitMonoTrueSilk,\n \"riff-16khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff16Khz16BitMonoPcm,\n \"riff-8khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff8Khz16BitMonoPcm,\n \"riff-24khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff24Khz16BitMonoPcm,\n \"riff-8khz-8bit-mono-mulaw\": SpeechSDK.SpeechSynthesisOutputFormat.Riff8Khz8BitMonoMULaw,\n \"raw-16khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw16Khz16BitMonoPcm,\n \"raw-24khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw24Khz16BitMonoPcm,\n \"raw-8khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw8Khz16BitMonoPcm,\n \"ogg-16khz-16bit-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Ogg16Khz16BitMonoOpus,\n \"ogg-24khz-16bit-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Ogg24Khz16BitMonoOpus,\n \"raw-48khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw48Khz16BitMonoPcm,\n \"riff-48khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff48Khz16BitMonoPcm,\n \"audio-48khz-96kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio48Khz96KBitRateMonoMp3,\n \"audio-48khz-192kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio48Khz192KBitRateMonoMp3,\n \"ogg-48khz-16bit-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Ogg48Khz16BitMonoOpus,\n \"webm-16khz-16bit-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Webm16Khz16BitMonoOpus,\n \"webm-24khz-16bit-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Webm24Khz16BitMonoOpus,\n \"webm-24khz-16bit-24kbps-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Webm24Khz16Bit24KbpsMonoOpus,\n \"raw-24khz-16bit-mono-truesilk\": SpeechSDK.SpeechSynthesisOutputFormat.Raw24Khz16BitMonoTrueSilk,\n \"raw-8khz-8bit-mono-alaw\": SpeechSDK.SpeechSynthesisOutputFormat.Raw8Khz8BitMonoALaw,\n \"riff-8khz-8bit-mono-alaw\": SpeechSDK.SpeechSynthesisOutputFormat.Riff8Khz8BitMonoALaw,\n \"audio-16khz-16bit-32kbps-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz16Bit32KbpsMonoOpus,\n \"audio-24khz-16bit-48kbps-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz16Bit48KbpsMonoOpus,\n \"audio-24khz-16bit-24kbps-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz16Bit24KbpsMonoOpus,\n \"raw-22050hz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw22050Hz16BitMonoPcm,\n \"riff-22050hz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff22050Hz16BitMonoPcm,\n \"raw-44100hz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw44100Hz16BitMonoPcm,\n \"riff-44100hz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff44100Hz16BitMonoPcm,\n \"amr-wb-16000hz\": SpeechSDK.SpeechSynthesisOutputFormat.AmrWb16000Hz,\n \"g722-16khz-64kbps\": SpeechSDK.SpeechSynthesisOutputFormat.G72216Khz64Kbps,\n};\n\nexport function resolveOutputFormat(outputFormat: string): SpeechSDK.SpeechSynthesisOutputFormat {\n const resolvedFormat = OUTPUT_FORMATS[outputFormat];\n if (resolvedFormat === undefined) {\n throw new Error(`Unsupported Azure Speech output format: ${outputFormat}`);\n }\n\n return resolvedFormat;\n}\n","import { synthesizeSpeech } from \"./synthesis.ts\";\nimport type { AzureTtsClientOptions } from \"./types.ts\";\n\nconst ENDPOINT_TEMPLATE = \"https://{region}.tts.speech.microsoft.com/cognitiveservices/v1\";\n\nexport class AzureTtsClient {\n readonly #options: AzureTtsClientOptions;\n\n constructor(options: AzureTtsClientOptions) {\n this.#options = options;\n }\n\n async synthesize(ssml: string): Promise<ArrayBuffer> {\n const { region, subscriptionKey, outputFormat, signal, timeoutMs } = this.#options;\n const endpoint = this.#options.endpoint?.trim() || ENDPOINT_TEMPLATE.replace(\"{region}\", region);\n this.#options.logger?.debug?.(\"Using Azure TTS endpoint:\", endpoint);\n\n const config = { endpoint, region, subscriptionKey, outputFormat, signal, timeoutMs };\n return synthesizeSpeech(ssml, config);\n }\n}\n"],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;;;ACAO,IAAM,gBAAN,cAA4B,MAAM;AAAA,EAMvC,YAAY,QAAgB,YAAoB,cAAsB,WAA0B;AAC9F,UAAM,6BAA6B,MAAM,IAAI,UAAU,EAAE;AACzD,SAAK,OAAO;AACZ,SAAK,SAAS;AACd,SAAK,aAAa;AAClB,SAAK,eAAe;AACpB,SAAK,YAAY;AAAA,EACnB;AACF;AAEO,IAAM,mBAAN,cAA+B,cAAc;AAAA,EAGlD,YAAY,cAAsB;AAChC,UAAM,GAAG,cAAc,cAAc,IAAI;AACzC,SAAK,OAAO;AACZ,SAAK,UAAU,+BAA+B,YAAY;AAC1D,SAAK,eAAe;AAAA,EACtB;AACF;AAEO,SAAS,qBAAqB,OAAkC;AACrE,QAAM,UAAU,iBAAiB,QAAQ,MAAM,UAAU,OAAO,KAAK;AACrE,SAAO,IAAI,iBAAiB,OAAO;AACrC;;;AC9BA,IAAAA,aAA2B;;;ACA3B,oDAA6B;;;ACA7B,gBAA2B;AAEpB,IAAM,wBAAwB;AAErC,IAAM,iBAAwE;AAAA,EAC5E,4BAAsC,sCAA4B;AAAA,EAClE,gCAA0C,sCAA4B;AAAA,EACtE,iCAA2C,sCAA4B;AAAA,EACvE,mCAA6C,sCAA4B;AAAA,EACzE,oCAA8C,sCAA4B;AAAA,EAC1E,mCAA6C,sCAA4B;AAAA,EACzE,mCAA6C,sCAA4B;AAAA,EACzE,mCAA6C,sCAA4B;AAAA,EACzE,oCAA8C,sCAA4B;AAAA,EAC1E,iCAA2C,sCAA4B;AAAA,EACvE,6BAAuC,sCAA4B;AAAA,EACnE,4BAAsC,sCAA4B;AAAA,EAClE,6BAAuC,sCAA4B;AAAA,EACnE,6BAAuC,sCAA4B;AAAA,EACnE,4BAAsC,sCAA4B;AAAA,EAClE,4BAAsC,sCAA4B;AAAA,EAClE,2BAAqC,sCAA4B;AAAA,EACjE,6BAAuC,sCAA4B;AAAA,EACnE,6BAAuC,sCAA4B;AAAA,EACnE,4BAAsC,sCAA4B;AAAA,EAClE,6BAAuC,sCAA4B;AAAA,EACnE,mCAA6C,sCAA4B;AAAA,EACzE,oCAA8C,sCAA4B;AAAA,EAC1E,6BAAuC,sCAA4B;AAAA,EACnE,8BAAwC,sCAA4B;AAAA,EACpE,8BAAwC,sCAA4B;AAAA,EACpE,qCAA+C,sCAA4B;AAAA,EAC3E,iCAA2C,sCAA4B;AAAA,EACvE,2BAAqC,sCAA4B;AAAA,EACjE,4BAAsC,sCAA4B;AAAA,EAClE,sCAAgD,sCAA4B;AAAA,EAC5E,sCAAgD,sCAA4B;AAAA,EAC5E,sCAAgD,sCAA4B;AAAA,EAC5E,8BAAwC,sCAA4B;AAAA,EACpE,+BAAyC,sCAA4B;AAAA,EACrE,8BAAwC,sCAA4B;AAAA,EACpE,+BAAyC,sCAA4B;AAAA,EACrE,kBAA4B,sCAA4B;AAAA,EACxD,qBAA+B,sCAA4B;AAC7D;AAEO,SAAS,oBAAoB,cAA6D;AAC/F,QAAM,iBAAiB,eAAe,YAAY;AAClD,MAAI,mBAAmB,QAAW;AAChC,UAAM,IAAI,MAAM,2CAA2C,YAAY,EAAE;AAAA,EAC3E;AAEA,SAAO;AACT;;;ADjDO,SAAS,gBAAgB,QAA2B;AACzD,QAAM,WAAW,OAAO,UAAU,KAAK,KAAK;AAC5C,SAAO,SAAS,QAAQ,eAAe,mBAAmB,OAAO,MAAM,CAAC;AAC1E;AAEO,SAAS,mBAAmB,QAAiC;AAClE,QAAM,EAAE,eAAe,uBAAuB,gBAAgB,IAAI;AAElE,QAAM,WAAW,IAAI,IAAI,gBAAgB,MAAM,CAAC;AAChD,QAAM,eAAe,2DAAa,aAAa,UAAU,eAAe;AACxE,eAAa,8BAA8B,oBAAoB,YAAY;AAC3E,SAAO;AACT;;;ADXA,SAAS,qBAAqB,cAAsC,aAAgD;AAClH,MAAI;AACF,gBAAY,MAAM;AAAA,EACpB,QAAQ;AAAA,EAAC;AAET,MAAI;AACF,iBAAa,MAAM;AAAA,EACrB,QAAQ;AAAA,EAAC;AACX;AAEA,eAAsB,iBAAiB,MAAc,QAAyC;AAC5F,MAAI,OAAO,QAAQ,SAAS;AAC1B,UAAM,qBAAqB,iCAAiC;AAAA,EAC9D;AAEA,QAAM,eAAe,mBAAmB,MAAM;AAC9C,QAAM,cAAc,IAAc,6BAAkB,cAAc,IAAI;AAEtE,SAAO,MAAM,IAAI,QAAqB,CAAC,SAAS,WAAW;AACzD,QAAI,kBAAkB;AACtB,QAAI,UAAU;AACd,QAAI;AACJ,QAAI;AACJ,UAAM,UAAU,MAAM;AACpB,UAAI,QAAS,cAAa,OAAO;AACjC,UAAI,aAAc,QAAO,QAAQ,oBAAoB,SAAS,YAAY;AAAA,IAC5E;AACA,UAAM,iBAAiB,MAAM;AAC3B,UAAI,gBAAiB;AACrB,wBAAkB;AAClB,2BAAqB,cAAc,WAAW;AAAA,IAChD;AACA,UAAM,kBAAkB,CAAC,UAAmB;AAC1C,UAAI,QAAS;AACb,gBAAU;AACV,cAAQ;AACR,qBAAe;AACf,aAAO,qBAAqB,KAAK,CAAC;AAAA,IACpC;AAEA,UAAM,KAAK,CAAC,WAA4C;AACtD,UAAI,QAAS;AACb,YAAM,EAAE,QAAQ,aAAa,IAAI;AACjC,UAAI,WAAqB,wBAAa,4BAA4B;AAChE,cAAM,MAAM,gBAAgB,uCAAuC,MAAM;AACzE,wBAAgB,GAAG;AACnB;AAAA,MACF;AACA,gBAAU;AACV,cAAQ;AACR,qBAAe;AACf,cAAQ,OAAO,SAAS;AAAA,IAC1B;AAEA,QAAI;AACF,UAAI,OAAO,QAAQ;AACjB,uBAAe,MAAM,gBAAgB,iCAAiC;AACtE,eAAO,OAAO,iBAAiB,SAAS,cAAc,EAAE,MAAM,KAAK,CAAC;AAAA,MACtE;AACA,UAAI,OAAO,cAAc,UAAa,OAAO,YAAY,GAAG;AAC1D,kBAAU;AAAA,UACR,MAAM,gBAAgB,oCAAoC,OAAO,SAAS,MAAM;AAAA,UAChF,OAAO;AAAA,QACT;AAAA,MACF;AACA,kBAAY,eAAe,MAAM,IAAI,eAAe;AAAA,IACtD,SAAS,OAAO;AACd,sBAAgB,KAAK;AAAA,IACvB;AAAA,EACF,CAAC;AACH;;;AGxEA,IAAM,oBAAoB;AAH1B;AAKO,IAAM,iBAAN,MAAqB;AAAA,EAG1B,YAAY,SAAgC;AAF5C,uBAAS;AAGP,uBAAK,UAAW;AAAA,EAClB;AAAA,EAEA,MAAM,WAAW,MAAoC;AACnD,UAAM,EAAE,QAAQ,iBAAiB,cAAc,QAAQ,UAAU,IAAI,mBAAK;AAC1E,UAAM,WAAW,mBAAK,UAAS,UAAU,KAAK,KAAK,kBAAkB,QAAQ,YAAY,MAAM;AAC/F,uBAAK,UAAS,QAAQ,QAAQ,6BAA6B,QAAQ;AAEnE,UAAM,SAAS,EAAE,UAAU,QAAQ,iBAAiB,cAAc,QAAQ,UAAU;AACpF,WAAO,iBAAiB,MAAM,MAAM;AAAA,EACtC;AACF;AAdW;","names":["SpeechSDK"]}
package/dist/index.mjs CHANGED
@@ -113,30 +113,56 @@ function closeSpeechResources(speechConfig, synthesizer) {
113
113
  }
114
114
  }
115
115
  async function synthesizeSpeech(ssml, config) {
116
+ if (config.signal?.aborted) {
117
+ throw createSpeechSdkError("Speech synthesis was cancelled.");
118
+ }
116
119
  const speechConfig = createSpeechConfig(config);
117
120
  const synthesizer = new SpeechSDK2.SpeechSynthesizer(speechConfig, null);
118
121
  return await new Promise((resolve, reject) => {
119
122
  let resourcesClosed = false;
123
+ let settled = false;
124
+ let timeout;
125
+ let abortHandler;
126
+ const cleanup = () => {
127
+ if (timeout) clearTimeout(timeout);
128
+ if (abortHandler) config.signal?.removeEventListener("abort", abortHandler);
129
+ };
120
130
  const closeResources = () => {
121
131
  if (resourcesClosed) return;
122
132
  resourcesClosed = true;
123
133
  closeSpeechResources(speechConfig, synthesizer);
124
134
  };
125
135
  const rejectWithError = (error) => {
136
+ if (settled) return;
137
+ settled = true;
138
+ cleanup();
126
139
  closeResources();
127
140
  reject(createSpeechSdkError(error));
128
141
  };
129
142
  const cb = (result) => {
143
+ if (settled) return;
130
144
  const { reason, errorDetails } = result;
131
145
  if (reason !== SpeechSDK2.ResultReason.SynthesizingAudioCompleted) {
132
146
  const err = errorDetails || `Speech synthesis failed with reason ${reason}.`;
133
147
  rejectWithError(err);
134
148
  return;
135
149
  }
150
+ settled = true;
151
+ cleanup();
136
152
  closeResources();
137
153
  resolve(result.audioData);
138
154
  };
139
155
  try {
156
+ if (config.signal) {
157
+ abortHandler = () => rejectWithError("Speech synthesis was cancelled.");
158
+ config.signal.addEventListener("abort", abortHandler, { once: true });
159
+ }
160
+ if (config.timeoutMs !== void 0 && config.timeoutMs > 0) {
161
+ timeout = setTimeout(
162
+ () => rejectWithError(`Speech synthesis timed out after ${config.timeoutMs} ms.`),
163
+ config.timeoutMs
164
+ );
165
+ }
140
166
  synthesizer.speakSsmlAsync(ssml, cb, rejectWithError);
141
167
  } catch (error) {
142
168
  rejectWithError(error);
@@ -153,10 +179,10 @@ var AzureTtsClient = class {
153
179
  __privateSet(this, _options, options);
154
180
  }
155
181
  async synthesize(ssml) {
156
- const { region, subscriptionKey, outputFormat } = __privateGet(this, _options);
182
+ const { region, subscriptionKey, outputFormat, signal, timeoutMs } = __privateGet(this, _options);
157
183
  const endpoint = __privateGet(this, _options).endpoint?.trim() || ENDPOINT_TEMPLATE.replace("{region}", region);
158
184
  __privateGet(this, _options).logger?.debug?.("Using Azure TTS endpoint:", endpoint);
159
- const config = { endpoint, region, subscriptionKey, outputFormat };
185
+ const config = { endpoint, region, subscriptionKey, outputFormat, signal, timeoutMs };
160
186
  return synthesizeSpeech(ssml, config);
161
187
  }
162
188
  };
package/dist/index.mjs.map CHANGED
@@ -1 +1 @@
1
- {"version":3,"sources":["../src/errors.ts","../src/synthesis.ts","../src/speechConfig.ts","../src/outputFormats.ts","../src/client.ts"],"sourcesContent":["export class AzureTtsError extends Error {\n readonly status: number;\n readonly statusText: string;\n readonly responseBody: string;\n readonly requestId: string | null;\n\n constructor(status: number, statusText: string, responseBody: string, requestId: string | null) {\n super(`Azure TTS request failed: ${status} ${statusText}`);\n this.name = \"AzureTtsError\";\n this.status = status;\n this.statusText = statusText;\n this.responseBody = responseBody;\n this.requestId = requestId;\n }\n}\n\nexport class AzureTtsSdkError extends AzureTtsError {\n readonly errorDetails: string;\n\n constructor(errorDetails: string) {\n super(0, \"Speech SDK\", errorDetails, null);\n this.name = \"AzureTtsSdkError\";\n this.message = `Azure TTS synthesis failed: ${errorDetails}`;\n this.errorDetails = errorDetails;\n }\n}\n\nexport function createSpeechSdkError(error: unknown): AzureTtsSdkError {\n const message = error instanceof Error ? error.message : String(error);\n return new AzureTtsSdkError(message);\n}\n","import * as SpeechSDK from \"microsoft-cognitiveservices-speech-sdk\";\nimport { createSpeechSdkError } from \"./errors.ts\";\nimport { createSpeechConfig } from \"./speechConfig.ts\";\nimport type { TtsConfig } from \"./types.ts\";\n\nfunction closeSpeechResources(speechConfig: SpeechSDK.SpeechConfig, synthesizer: SpeechSDK.SpeechSynthesizer): void {\n try {\n synthesizer.close();\n } catch {}\n\n try {\n speechConfig.close();\n } catch {}\n}\n\nexport async function synthesizeSpeech(ssml: string, config: TtsConfig): Promise<ArrayBuffer> {\n const speechConfig = createSpeechConfig(config);\n const synthesizer = new SpeechSDK.SpeechSynthesizer(speechConfig, null);\n\n return await new Promise<ArrayBuffer>((resolve, reject) => {\n let resourcesClosed = false;\n const closeResources = () => {\n if (resourcesClosed) return;\n resourcesClosed = true;\n closeSpeechResources(speechConfig, synthesizer);\n };\n const rejectWithError = (error: unknown) => {\n closeResources();\n reject(createSpeechSdkError(error));\n };\n\n const cb = (result: SpeechSDK.SpeechSynthesisResult) => {\n const { reason, errorDetails } = result;\n if (reason !== SpeechSDK.ResultReason.SynthesizingAudioCompleted) {\n const err = errorDetails || `Speech synthesis failed with reason ${reason}.`;\n rejectWithError(err);\n return;\n }\n closeResources();\n resolve(result.audioData);\n };\n\n try {\n synthesizer.speakSsmlAsync(ssml, cb, rejectWithError);\n } catch (error) {\n rejectWithError(error);\n }\n });\n}\n","import { SpeechConfig } from \"microsoft-cognitiveservices-speech-sdk\";\nimport { DEFAULT_OUTPUT_FORMAT, resolveOutputFormat } from \"./outputFormats.ts\";\nimport type { TtsConfig } from \"./types.ts\";\n\nexport function resolveEndpoint(config: TtsConfig): string {\n const endpoint = config.endpoint?.trim() || \"https://{region}.tts.speech.microsoft.com/cognitiveservices/v1\";\n return endpoint.replace(/\\{region\\}/g, encodeURIComponent(config.region));\n}\n\nexport function createSpeechConfig(config: TtsConfig): SpeechConfig {\n const { outputFormat = DEFAULT_OUTPUT_FORMAT, subscriptionKey } = config;\n\n const endpoint = new URL(resolveEndpoint(config));\n const speechConfig = SpeechConfig.fromEndpoint(endpoint, subscriptionKey);\n speechConfig.speechSynthesisOutputFormat = resolveOutputFormat(outputFormat);\n return speechConfig;\n}\n","import * as SpeechSDK from \"microsoft-cognitiveservices-speech-sdk\";\n\nexport const DEFAULT_OUTPUT_FORMAT = \"audio-16khz-128kbitrate-mono-mp3\";\n\nconst OUTPUT_FORMATS: Record<string, SpeechSDK.SpeechSynthesisOutputFormat> = {\n \"raw-8khz-8bit-mono-mulaw\": SpeechSDK.SpeechSynthesisOutputFormat.Raw8Khz8BitMonoMULaw,\n \"riff-16khz-16kbps-mono-siren\": SpeechSDK.SpeechSynthesisOutputFormat.Riff16Khz16KbpsMonoSiren,\n \"audio-16khz-16kbps-mono-siren\": SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz16KbpsMonoSiren,\n \"audio-16khz-32kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz32KBitRateMonoMp3,\n \"audio-16khz-128kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz128KBitRateMonoMp3,\n \"audio-16khz-64kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz64KBitRateMonoMp3,\n \"audio-24khz-48kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz48KBitRateMonoMp3,\n \"audio-24khz-96kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz96KBitRateMonoMp3,\n \"audio-24khz-160kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz160KBitRateMonoMp3,\n \"raw-16khz-16bit-mono-truesilk\": SpeechSDK.SpeechSynthesisOutputFormat.Raw16Khz16BitMonoTrueSilk,\n \"riff-16khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff16Khz16BitMonoPcm,\n \"riff-8khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff8Khz16BitMonoPcm,\n \"riff-24khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff24Khz16BitMonoPcm,\n \"riff-8khz-8bit-mono-mulaw\": SpeechSDK.SpeechSynthesisOutputFormat.Riff8Khz8BitMonoMULaw,\n \"raw-16khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw16Khz16BitMonoPcm,\n \"raw-24khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw24Khz16BitMonoPcm,\n \"raw-8khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw8Khz16BitMonoPcm,\n \"ogg-16khz-16bit-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Ogg16Khz16BitMonoOpus,\n \"ogg-24khz-16bit-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Ogg24Khz16BitMonoOpus,\n \"raw-48khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw48Khz16BitMonoPcm,\n \"riff-48khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff48Khz16BitMonoPcm,\n \"audio-48khz-96kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio48Khz96KBitRateMonoMp3,\n \"audio-48khz-192kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio48Khz192KBitRateMonoMp3,\n \"ogg-48khz-16bit-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Ogg48Khz16BitMonoOpus,\n \"webm-16khz-16bit-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Webm16Khz16BitMonoOpus,\n \"webm-24khz-16bit-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Webm24Khz16BitMonoOpus,\n \"webm-24khz-16bit-24kbps-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Webm24Khz16Bit24KbpsMonoOpus,\n \"raw-24khz-16bit-mono-truesilk\": SpeechSDK.SpeechSynthesisOutputFormat.Raw24Khz16BitMonoTrueSilk,\n \"raw-8khz-8bit-mono-alaw\": SpeechSDK.SpeechSynthesisOutputFormat.Raw8Khz8BitMonoALaw,\n \"riff-8khz-8bit-mono-alaw\": SpeechSDK.SpeechSynthesisOutputFormat.Riff8Khz8BitMonoALaw,\n \"audio-16khz-16bit-32kbps-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz16Bit32KbpsMonoOpus,\n \"audio-24khz-16bit-48kbps-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz16Bit48KbpsMonoOpus,\n \"audio-24khz-16bit-24kbps-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz16Bit24KbpsMonoOpus,\n \"raw-22050hz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw22050Hz16BitMonoPcm,\n \"riff-22050hz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff22050Hz16BitMonoPcm,\n \"raw-44100hz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw44100Hz16BitMonoPcm,\n \"riff-44100hz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff44100Hz16BitMonoPcm,\n \"amr-wb-16000hz\": SpeechSDK.SpeechSynthesisOutputFormat.AmrWb16000Hz,\n \"g722-16khz-64kbps\": SpeechSDK.SpeechSynthesisOutputFormat.G72216Khz64Kbps,\n};\n\nexport function resolveOutputFormat(outputFormat: string): SpeechSDK.SpeechSynthesisOutputFormat {\n const resolvedFormat = OUTPUT_FORMATS[outputFormat];\n if (resolvedFormat === undefined) {\n throw new Error(`Unsupported Azure Speech output format: ${outputFormat}`);\n }\n\n return resolvedFormat;\n}\n","import { synthesizeSpeech } from \"./synthesis.ts\";\nimport type { AzureTtsClientOptions } from \"./types.ts\";\n\nconst ENDPOINT_TEMPLATE = \"https://{region}.tts.speech.microsoft.com/cognitiveservices/v1\";\n\nexport class AzureTtsClient {\n readonly #options: AzureTtsClientOptions;\n\n constructor(options: AzureTtsClientOptions) {\n this.#options = options;\n }\n\n async synthesize(ssml: string): Promise<ArrayBuffer> {\n const { region, subscriptionKey, outputFormat } = this.#options;\n const endpoint = this.#options.endpoint?.trim() || ENDPOINT_TEMPLATE.replace(\"{region}\", region);\n this.#options.logger?.debug?.(\"Using Azure TTS endpoint:\", endpoint);\n\n const config = { endpoint, region, subscriptionKey, outputFormat };\n return synthesizeSpeech(ssml, config);\n }\n}\n"],"mappings":";;;;;;;;;AAAO,IAAM,gBAAN,cAA4B,MAAM;AAAA,EAMvC,YAAY,QAAgB,YAAoB,cAAsB,WAA0B;AAC9F,UAAM,6BAA6B,MAAM,IAAI,UAAU,EAAE;AACzD,SAAK,OAAO;AACZ,SAAK,SAAS;AACd,SAAK,aAAa;AAClB,SAAK,eAAe;AACpB,SAAK,YAAY;AAAA,EACnB;AACF;AAEO,IAAM,mBAAN,cAA+B,cAAc;AAAA,EAGlD,YAAY,cAAsB;AAChC,UAAM,GAAG,cAAc,cAAc,IAAI;AACzC,SAAK,OAAO;AACZ,SAAK,UAAU,+BAA+B,YAAY;AAC1D,SAAK,eAAe;AAAA,EACtB;AACF;AAEO,SAAS,qBAAqB,OAAkC;AACrE,QAAM,UAAU,iBAAiB,QAAQ,MAAM,UAAU,OAAO,KAAK;AACrE,SAAO,IAAI,iBAAiB,OAAO;AACrC;;;AC9BA,YAAYA,gBAAe;;;ACA3B,SAAS,oBAAoB;;;ACA7B,YAAY,eAAe;AAEpB,IAAM,wBAAwB;AAErC,IAAM,iBAAwE;AAAA,EAC5E,4BAAsC,sCAA4B;AAAA,EAClE,gCAA0C,sCAA4B;AAAA,EACtE,iCAA2C,sCAA4B;AAAA,EACvE,mCAA6C,sCAA4B;AAAA,EACzE,oCAA8C,sCAA4B;AAAA,EAC1E,mCAA6C,sCAA4B;AAAA,EACzE,mCAA6C,sCAA4B;AAAA,EACzE,mCAA6C,sCAA4B;AAAA,EACzE,oCAA8C,sCAA4B;AAAA,EAC1E,iCAA2C,sCAA4B;AAAA,EACvE,6BAAuC,sCAA4B;AAAA,EACnE,4BAAsC,sCAA4B;AAAA,EAClE,6BAAuC,sCAA4B;AAAA,EACnE,6BAAuC,sCAA4B;AAAA,EACnE,4BAAsC,sCAA4B;AAAA,EAClE,4BAAsC,sCAA4B;AAAA,EAClE,2BAAqC,sCAA4B;AAAA,EACjE,6BAAuC,sCAA4B;AAAA,EACnE,6BAAuC,sCAA4B;AAAA,EACnE,4BAAsC,sCAA4B;AAAA,EAClE,6BAAuC,sCAA4B;AAAA,EACnE,mCAA6C,sCAA4B;AAAA,EACzE,oCAA8C,sCAA4B;AAAA,EAC1E,6BAAuC,sCAA4B;AAAA,EACnE,8BAAwC,sCAA4B;AAAA,EACpE,8BAAwC,sCAA4B;AAAA,EACpE,qCAA+C,sCAA4B;AAAA,EAC3E,iCAA2C,sCAA4B;AAAA,EACvE,2BAAqC,sCAA4B;AAAA,EACjE,4BAAsC,sCAA4B;AAAA,EAClE,sCAAgD,sCAA4B;AAAA,EAC5E,sCAAgD,sCAA4B;AAAA,EAC5E,sCAAgD,sCAA4B;AAAA,EAC5E,8BAAwC,sCAA4B;AAAA,EACpE,+BAAyC,sCAA4B;AAAA,EACrE,8BAAwC,sCAA4B;AAAA,EACpE,+BAAyC,sCAA4B;AAAA,EACrE,kBAA4B,sCAA4B;AAAA,EACxD,qBAA+B,sCAA4B;AAC7D;AAEO,SAAS,oBAAoB,cAA6D;AAC/F,QAAM,iBAAiB,eAAe,YAAY;AAClD,MAAI,mBAAmB,QAAW;AAChC,UAAM,IAAI,MAAM,2CAA2C,YAAY,EAAE;AAAA,EAC3E;AAEA,SAAO;AACT;;;ADjDO,SAAS,gBAAgB,QAA2B;AACzD,QAAM,WAAW,OAAO,UAAU,KAAK,KAAK;AAC5C,SAAO,SAAS,QAAQ,eAAe,mBAAmB,OAAO,MAAM,CAAC;AAC1E;AAEO,SAAS,mBAAmB,QAAiC;AAClE,QAAM,EAAE,eAAe,uBAAuB,gBAAgB,IAAI;AAElE,QAAM,WAAW,IAAI,IAAI,gBAAgB,MAAM,CAAC;AAChD,QAAM,eAAe,aAAa,aAAa,UAAU,eAAe;AACxE,eAAa,8BAA8B,oBAAoB,YAAY;AAC3E,SAAO;AACT;;;ADXA,SAAS,qBAAqB,cAAsC,aAAgD;AAClH,MAAI;AACF,gBAAY,MAAM;AAAA,EACpB,QAAQ;AAAA,EAAC;AAET,MAAI;AACF,iBAAa,MAAM;AAAA,EACrB,QAAQ;AAAA,EAAC;AACX;AAEA,eAAsB,iBAAiB,MAAc,QAAyC;AAC5F,QAAM,eAAe,mBAAmB,MAAM;AAC9C,QAAM,cAAc,IAAc,6BAAkB,cAAc,IAAI;AAEtE,SAAO,MAAM,IAAI,QAAqB,CAAC,SAAS,WAAW;AACzD,QAAI,kBAAkB;AACtB,UAAM,iBAAiB,MAAM;AAC3B,UAAI,gBAAiB;AACrB,wBAAkB;AAClB,2BAAqB,cAAc,WAAW;AAAA,IAChD;AACA,UAAM,kBAAkB,CAAC,UAAmB;AAC1C,qBAAe;AACf,aAAO,qBAAqB,KAAK,CAAC;AAAA,IACpC;AAEA,UAAM,KAAK,CAAC,WAA4C;AACtD,YAAM,EAAE,QAAQ,aAAa,IAAI;AACjC,UAAI,WAAqB,wBAAa,4BAA4B;AAChE,cAAM,MAAM,gBAAgB,uCAAuC,MAAM;AACzE,wBAAgB,GAAG;AACnB;AAAA,MACF;AACA,qBAAe;AACf,cAAQ,OAAO,SAAS;AAAA,IAC1B;AAEA,QAAI;AACF,kBAAY,eAAe,MAAM,IAAI,eAAe;AAAA,IACtD,SAAS,OAAO;AACd,sBAAgB,KAAK;AAAA,IACvB;AAAA,EACF,CAAC;AACH;;;AG7CA,IAAM,oBAAoB;AAH1B;AAKO,IAAM,iBAAN,MAAqB;AAAA,EAG1B,YAAY,SAAgC;AAF5C,uBAAS;AAGP,uBAAK,UAAW;AAAA,EAClB;AAAA,EAEA,MAAM,WAAW,MAAoC;AACnD,UAAM,EAAE,QAAQ,iBAAiB,aAAa,IAAI,mBAAK;AACvD,UAAM,WAAW,mBAAK,UAAS,UAAU,KAAK,KAAK,kBAAkB,QAAQ,YAAY,MAAM;AAC/F,uBAAK,UAAS,QAAQ,QAAQ,6BAA6B,QAAQ;AAEnE,UAAM,SAAS,EAAE,UAAU,QAAQ,iBAAiB,aAAa;AACjE,WAAO,iBAAiB,MAAM,MAAM;AAAA,EACtC;AACF;AAdW;","names":["SpeechSDK"]}
1
+ {"version":3,"sources":["../src/errors.ts","../src/synthesis.ts","../src/speechConfig.ts","../src/outputFormats.ts","../src/client.ts"],"sourcesContent":["export class AzureTtsError extends Error {\n readonly status: number;\n readonly statusText: string;\n readonly responseBody: string;\n readonly requestId: string | null;\n\n constructor(status: number, statusText: string, responseBody: string, requestId: string | null) {\n super(`Azure TTS request failed: ${status} ${statusText}`);\n this.name = \"AzureTtsError\";\n this.status = status;\n this.statusText = statusText;\n this.responseBody = responseBody;\n this.requestId = requestId;\n }\n}\n\nexport class AzureTtsSdkError extends AzureTtsError {\n readonly errorDetails: string;\n\n constructor(errorDetails: string) {\n super(0, \"Speech SDK\", errorDetails, null);\n this.name = \"AzureTtsSdkError\";\n this.message = `Azure TTS synthesis failed: ${errorDetails}`;\n this.errorDetails = errorDetails;\n }\n}\n\nexport function createSpeechSdkError(error: unknown): AzureTtsSdkError {\n const message = error instanceof Error ? error.message : String(error);\n return new AzureTtsSdkError(message);\n}\n","import * as SpeechSDK from \"microsoft-cognitiveservices-speech-sdk\";\nimport { createSpeechSdkError } from \"./errors.ts\";\nimport { createSpeechConfig } from \"./speechConfig.ts\";\nimport type { TtsConfig } from \"./types.ts\";\n\nfunction closeSpeechResources(speechConfig: SpeechSDK.SpeechConfig, synthesizer: SpeechSDK.SpeechSynthesizer): void {\n try {\n synthesizer.close();\n } catch {}\n\n try {\n speechConfig.close();\n } catch {}\n}\n\nexport async function synthesizeSpeech(ssml: string, config: TtsConfig): Promise<ArrayBuffer> {\n if (config.signal?.aborted) {\n throw createSpeechSdkError(\"Speech synthesis was cancelled.\");\n }\n\n const speechConfig = createSpeechConfig(config);\n const synthesizer = new SpeechSDK.SpeechSynthesizer(speechConfig, null);\n\n return await new Promise<ArrayBuffer>((resolve, reject) => {\n let resourcesClosed = false;\n let settled = false;\n let timeout: ReturnType<typeof setTimeout> | undefined;\n let abortHandler: (() => void) | undefined;\n const cleanup = () => {\n if (timeout) clearTimeout(timeout);\n if (abortHandler) config.signal?.removeEventListener(\"abort\", abortHandler);\n };\n const closeResources = () => {\n if (resourcesClosed) return;\n resourcesClosed = true;\n closeSpeechResources(speechConfig, synthesizer);\n };\n const rejectWithError = (error: unknown) => {\n if (settled) return;\n settled = true;\n cleanup();\n closeResources();\n reject(createSpeechSdkError(error));\n };\n\n const cb = (result: SpeechSDK.SpeechSynthesisResult) => {\n if (settled) return;\n const { reason, errorDetails } = result;\n if (reason !== SpeechSDK.ResultReason.SynthesizingAudioCompleted) {\n const err = errorDetails || `Speech synthesis failed with reason ${reason}.`;\n rejectWithError(err);\n return;\n }\n settled = true;\n cleanup();\n closeResources();\n resolve(result.audioData);\n };\n\n try {\n if (config.signal) {\n abortHandler = () => rejectWithError(\"Speech synthesis was cancelled.\");\n config.signal.addEventListener(\"abort\", abortHandler, { once: true });\n }\n if (config.timeoutMs !== undefined && config.timeoutMs > 0) {\n timeout = setTimeout(\n () => rejectWithError(`Speech synthesis timed out after ${config.timeoutMs} ms.`),\n config.timeoutMs,\n );\n }\n synthesizer.speakSsmlAsync(ssml, cb, rejectWithError);\n } catch (error) {\n rejectWithError(error);\n }\n });\n}\n","import { SpeechConfig } from \"microsoft-cognitiveservices-speech-sdk\";\nimport { DEFAULT_OUTPUT_FORMAT, resolveOutputFormat } from \"./outputFormats.ts\";\nimport type { TtsConfig } from \"./types.ts\";\n\nexport function resolveEndpoint(config: TtsConfig): string {\n const endpoint = config.endpoint?.trim() || \"https://{region}.tts.speech.microsoft.com/cognitiveservices/v1\";\n return endpoint.replace(/\\{region\\}/g, encodeURIComponent(config.region));\n}\n\nexport function createSpeechConfig(config: TtsConfig): SpeechConfig {\n const { outputFormat = DEFAULT_OUTPUT_FORMAT, subscriptionKey } = config;\n\n const endpoint = new URL(resolveEndpoint(config));\n const speechConfig = SpeechConfig.fromEndpoint(endpoint, subscriptionKey);\n speechConfig.speechSynthesisOutputFormat = resolveOutputFormat(outputFormat);\n return speechConfig;\n}\n","import * as SpeechSDK from \"microsoft-cognitiveservices-speech-sdk\";\n\nexport const DEFAULT_OUTPUT_FORMAT = \"audio-16khz-128kbitrate-mono-mp3\";\n\nconst OUTPUT_FORMATS: Record<string, SpeechSDK.SpeechSynthesisOutputFormat> = {\n \"raw-8khz-8bit-mono-mulaw\": SpeechSDK.SpeechSynthesisOutputFormat.Raw8Khz8BitMonoMULaw,\n \"riff-16khz-16kbps-mono-siren\": SpeechSDK.SpeechSynthesisOutputFormat.Riff16Khz16KbpsMonoSiren,\n \"audio-16khz-16kbps-mono-siren\": SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz16KbpsMonoSiren,\n \"audio-16khz-32kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz32KBitRateMonoMp3,\n \"audio-16khz-128kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz128KBitRateMonoMp3,\n \"audio-16khz-64kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz64KBitRateMonoMp3,\n \"audio-24khz-48kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz48KBitRateMonoMp3,\n \"audio-24khz-96kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz96KBitRateMonoMp3,\n \"audio-24khz-160kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz160KBitRateMonoMp3,\n \"raw-16khz-16bit-mono-truesilk\": SpeechSDK.SpeechSynthesisOutputFormat.Raw16Khz16BitMonoTrueSilk,\n \"riff-16khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff16Khz16BitMonoPcm,\n \"riff-8khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff8Khz16BitMonoPcm,\n \"riff-24khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff24Khz16BitMonoPcm,\n \"riff-8khz-8bit-mono-mulaw\": SpeechSDK.SpeechSynthesisOutputFormat.Riff8Khz8BitMonoMULaw,\n \"raw-16khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw16Khz16BitMonoPcm,\n \"raw-24khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw24Khz16BitMonoPcm,\n \"raw-8khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw8Khz16BitMonoPcm,\n \"ogg-16khz-16bit-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Ogg16Khz16BitMonoOpus,\n \"ogg-24khz-16bit-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Ogg24Khz16BitMonoOpus,\n \"raw-48khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw48Khz16BitMonoPcm,\n \"riff-48khz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff48Khz16BitMonoPcm,\n \"audio-48khz-96kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio48Khz96KBitRateMonoMp3,\n \"audio-48khz-192kbitrate-mono-mp3\": SpeechSDK.SpeechSynthesisOutputFormat.Audio48Khz192KBitRateMonoMp3,\n \"ogg-48khz-16bit-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Ogg48Khz16BitMonoOpus,\n \"webm-16khz-16bit-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Webm16Khz16BitMonoOpus,\n \"webm-24khz-16bit-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Webm24Khz16BitMonoOpus,\n \"webm-24khz-16bit-24kbps-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Webm24Khz16Bit24KbpsMonoOpus,\n \"raw-24khz-16bit-mono-truesilk\": SpeechSDK.SpeechSynthesisOutputFormat.Raw24Khz16BitMonoTrueSilk,\n \"raw-8khz-8bit-mono-alaw\": SpeechSDK.SpeechSynthesisOutputFormat.Raw8Khz8BitMonoALaw,\n \"riff-8khz-8bit-mono-alaw\": SpeechSDK.SpeechSynthesisOutputFormat.Riff8Khz8BitMonoALaw,\n \"audio-16khz-16bit-32kbps-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz16Bit32KbpsMonoOpus,\n \"audio-24khz-16bit-48kbps-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz16Bit48KbpsMonoOpus,\n \"audio-24khz-16bit-24kbps-mono-opus\": SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz16Bit24KbpsMonoOpus,\n \"raw-22050hz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw22050Hz16BitMonoPcm,\n \"riff-22050hz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff22050Hz16BitMonoPcm,\n \"raw-44100hz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Raw44100Hz16BitMonoPcm,\n \"riff-44100hz-16bit-mono-pcm\": SpeechSDK.SpeechSynthesisOutputFormat.Riff44100Hz16BitMonoPcm,\n \"amr-wb-16000hz\": SpeechSDK.SpeechSynthesisOutputFormat.AmrWb16000Hz,\n \"g722-16khz-64kbps\": SpeechSDK.SpeechSynthesisOutputFormat.G72216Khz64Kbps,\n};\n\nexport function resolveOutputFormat(outputFormat: string): SpeechSDK.SpeechSynthesisOutputFormat {\n const resolvedFormat = OUTPUT_FORMATS[outputFormat];\n if (resolvedFormat === undefined) {\n throw new Error(`Unsupported Azure Speech output format: ${outputFormat}`);\n }\n\n return resolvedFormat;\n}\n","import { synthesizeSpeech } from \"./synthesis.ts\";\nimport type { AzureTtsClientOptions } from \"./types.ts\";\n\nconst ENDPOINT_TEMPLATE = \"https://{region}.tts.speech.microsoft.com/cognitiveservices/v1\";\n\nexport class AzureTtsClient {\n readonly #options: AzureTtsClientOptions;\n\n constructor(options: AzureTtsClientOptions) {\n this.#options = options;\n }\n\n async synthesize(ssml: string): Promise<ArrayBuffer> {\n const { region, subscriptionKey, outputFormat, signal, timeoutMs } = this.#options;\n const endpoint = this.#options.endpoint?.trim() || ENDPOINT_TEMPLATE.replace(\"{region}\", region);\n this.#options.logger?.debug?.(\"Using Azure TTS endpoint:\", endpoint);\n\n const config = { endpoint, region, subscriptionKey, outputFormat, signal, timeoutMs };\n return synthesizeSpeech(ssml, config);\n }\n}\n"],"mappings":";;;;;;;;;AAAO,IAAM,gBAAN,cAA4B,MAAM;AAAA,EAMvC,YAAY,QAAgB,YAAoB,cAAsB,WAA0B;AAC9F,UAAM,6BAA6B,MAAM,IAAI,UAAU,EAAE;AACzD,SAAK,OAAO;AACZ,SAAK,SAAS;AACd,SAAK,aAAa;AAClB,SAAK,eAAe;AACpB,SAAK,YAAY;AAAA,EACnB;AACF;AAEO,IAAM,mBAAN,cAA+B,cAAc;AAAA,EAGlD,YAAY,cAAsB;AAChC,UAAM,GAAG,cAAc,cAAc,IAAI;AACzC,SAAK,OAAO;AACZ,SAAK,UAAU,+BAA+B,YAAY;AAC1D,SAAK,eAAe;AAAA,EACtB;AACF;AAEO,SAAS,qBAAqB,OAAkC;AACrE,QAAM,UAAU,iBAAiB,QAAQ,MAAM,UAAU,OAAO,KAAK;AACrE,SAAO,IAAI,iBAAiB,OAAO;AACrC;;;AC9BA,YAAYA,gBAAe;;;ACA3B,SAAS,oBAAoB;;;ACA7B,YAAY,eAAe;AAEpB,IAAM,wBAAwB;AAErC,IAAM,iBAAwE;AAAA,EAC5E,4BAAsC,sCAA4B;AAAA,EAClE,gCAA0C,sCAA4B;AAAA,EACtE,iCAA2C,sCAA4B;AAAA,EACvE,mCAA6C,sCAA4B;AAAA,EACzE,oCAA8C,sCAA4B;AAAA,EAC1E,mCAA6C,sCAA4B;AAAA,EACzE,mCAA6C,sCAA4B;AAAA,EACzE,mCAA6C,sCAA4B;AAAA,EACzE,oCAA8C,sCAA4B;AAAA,EAC1E,iCAA2C,sCAA4B;AAAA,EACvE,6BAAuC,sCAA4B;AAAA,EACnE,4BAAsC,sCAA4B;AAAA,EAClE,6BAAuC,sCAA4B;AAAA,EACnE,6BAAuC,sCAA4B;AAAA,EACnE,4BAAsC,sCAA4B;AAAA,EAClE,4BAAsC,sCAA4B;AAAA,EAClE,2BAAqC,sCAA4B;AAAA,EACjE,6BAAuC,sCAA4B;AAAA,EACnE,6BAAuC,sCAA4B;AAAA,EACnE,4BAAsC,sCAA4B;AAAA,EAClE,6BAAuC,sCAA4B;AAAA,EACnE,mCAA6C,sCAA4B;AAAA,EACzE,oCAA8C,sCAA4B;AAAA,EAC1E,6BAAuC,sCAA4B;AAAA,EACnE,8BAAwC,sCAA4B;AAAA,EACpE,8BAAwC,sCAA4B;AAAA,EACpE,qCAA+C,sCAA4B;AAAA,EAC3E,iCAA2C,sCAA4B;AAAA,EACvE,2BAAqC,sCAA4B;AAAA,EACjE,4BAAsC,sCAA4B;AAAA,EAClE,sCAAgD,sCAA4B;AAAA,EAC5E,sCAAgD,sCAA4B;AAAA,EAC5E,sCAAgD,sCAA4B;AAAA,EAC5E,8BAAwC,sCAA4B;AAAA,EACpE,+BAAyC,sCAA4B;AAAA,EACrE,8BAAwC,sCAA4B;AAAA,EACpE,+BAAyC,sCAA4B;AAAA,EACrE,kBAA4B,sCAA4B;AAAA,EACxD,qBAA+B,sCAA4B;AAC7D;AAEO,SAAS,oBAAoB,cAA6D;AAC/F,QAAM,iBAAiB,eAAe,YAAY;AAClD,MAAI,mBAAmB,QAAW;AAChC,UAAM,IAAI,MAAM,2CAA2C,YAAY,EAAE;AAAA,EAC3E;AAEA,SAAO;AACT;;;ADjDO,SAAS,gBAAgB,QAA2B;AACzD,QAAM,WAAW,OAAO,UAAU,KAAK,KAAK;AAC5C,SAAO,SAAS,QAAQ,eAAe,mBAAmB,OAAO,MAAM,CAAC;AAC1E;AAEO,SAAS,mBAAmB,QAAiC;AAClE,QAAM,EAAE,eAAe,uBAAuB,gBAAgB,IAAI;AAElE,QAAM,WAAW,IAAI,IAAI,gBAAgB,MAAM,CAAC;AAChD,QAAM,eAAe,aAAa,aAAa,UAAU,eAAe;AACxE,eAAa,8BAA8B,oBAAoB,YAAY;AAC3E,SAAO;AACT;;;ADXA,SAAS,qBAAqB,cAAsC,aAAgD;AAClH,MAAI;AACF,gBAAY,MAAM;AAAA,EACpB,QAAQ;AAAA,EAAC;AAET,MAAI;AACF,iBAAa,MAAM;AAAA,EACrB,QAAQ;AAAA,EAAC;AACX;AAEA,eAAsB,iBAAiB,MAAc,QAAyC;AAC5F,MAAI,OAAO,QAAQ,SAAS;AAC1B,UAAM,qBAAqB,iCAAiC;AAAA,EAC9D;AAEA,QAAM,eAAe,mBAAmB,MAAM;AAC9C,QAAM,cAAc,IAAc,6BAAkB,cAAc,IAAI;AAEtE,SAAO,MAAM,IAAI,QAAqB,CAAC,SAAS,WAAW;AACzD,QAAI,kBAAkB;AACtB,QAAI,UAAU;AACd,QAAI;AACJ,QAAI;AACJ,UAAM,UAAU,MAAM;AACpB,UAAI,QAAS,cAAa,OAAO;AACjC,UAAI,aAAc,QAAO,QAAQ,oBAAoB,SAAS,YAAY;AAAA,IAC5E;AACA,UAAM,iBAAiB,MAAM;AAC3B,UAAI,gBAAiB;AACrB,wBAAkB;AAClB,2BAAqB,cAAc,WAAW;AAAA,IAChD;AACA,UAAM,kBAAkB,CAAC,UAAmB;AAC1C,UAAI,QAAS;AACb,gBAAU;AACV,cAAQ;AACR,qBAAe;AACf,aAAO,qBAAqB,KAAK,CAAC;AAAA,IACpC;AAEA,UAAM,KAAK,CAAC,WAA4C;AACtD,UAAI,QAAS;AACb,YAAM,EAAE,QAAQ,aAAa,IAAI;AACjC,UAAI,WAAqB,wBAAa,4BAA4B;AAChE,cAAM,MAAM,gBAAgB,uCAAuC,MAAM;AACzE,wBAAgB,GAAG;AACnB;AAAA,MACF;AACA,gBAAU;AACV,cAAQ;AACR,qBAAe;AACf,cAAQ,OAAO,SAAS;AAAA,IAC1B;AAEA,QAAI;AACF,UAAI,OAAO,QAAQ;AACjB,uBAAe,MAAM,gBAAgB,iCAAiC;AACtE,eAAO,OAAO,iBAAiB,SAAS,cAAc,EAAE,MAAM,KAAK,CAAC;AAAA,MACtE;AACA,UAAI,OAAO,cAAc,UAAa,OAAO,YAAY,GAAG;AAC1D,kBAAU;AAAA,UACR,MAAM,gBAAgB,oCAAoC,OAAO,SAAS,MAAM;AAAA,UAChF,OAAO;AAAA,QACT;AAAA,MACF;AACA,kBAAY,eAAe,MAAM,IAAI,eAAe;AAAA,IACtD,SAAS,OAAO;AACd,sBAAgB,KAAK;AAAA,IACvB;AAAA,EACF,CAAC;AACH;;;AGxEA,IAAM,oBAAoB;AAH1B;AAKO,IAAM,iBAAN,MAAqB;AAAA,EAG1B,YAAY,SAAgC;AAF5C,uBAAS;AAGP,uBAAK,UAAW;AAAA,EAClB;AAAA,EAEA,MAAM,WAAW,MAAoC;AACnD,UAAM,EAAE,QAAQ,iBAAiB,cAAc,QAAQ,UAAU,IAAI,mBAAK;AAC1E,UAAM,WAAW,mBAAK,UAAS,UAAU,KAAK,KAAK,kBAAkB,QAAQ,YAAY,MAAM;AAC/F,uBAAK,UAAS,QAAQ,QAAQ,6BAA6B,QAAQ;AAEnE,UAAM,SAAS,EAAE,UAAU,QAAQ,iBAAiB,cAAc,QAAQ,UAAU;AACpF,WAAO,iBAAiB,MAAM,MAAM;AAAA,EACtC;AACF;AAdW;","names":["SpeechSDK"]}
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@ssml-builder-js/azure-tts-client",
3
- "version": "2.4.0",
3
+ "version": "2.6.0",
4
4
  "description": "Azure Text-to-Speech client using the Microsoft Speech SDK",
5
5
  "license": "MIT",
6
6
  "repository": {
package/src/client.ts CHANGED
@@ -11,11 +11,11 @@ export class AzureTtsClient {
11
11
  }
12
12
 
13
13
  async synthesize(ssml: string): Promise<ArrayBuffer> {
14
- const { region, subscriptionKey, outputFormat } = this.#options;
14
+ const { region, subscriptionKey, outputFormat, signal, timeoutMs } = this.#options;
15
15
  const endpoint = this.#options.endpoint?.trim() || ENDPOINT_TEMPLATE.replace("{region}", region);
16
16
  this.#options.logger?.debug?.("Using Azure TTS endpoint:", endpoint);
17
17
 
18
- const config = { endpoint, region, subscriptionKey, outputFormat };
18
+ const config = { endpoint, region, subscriptionKey, outputFormat, signal, timeoutMs };
19
19
  return synthesizeSpeech(ssml, config);
20
20
  }
21
21
  }
package/src/synthesis.ts CHANGED
@@ -14,33 +14,60 @@ function closeSpeechResources(speechConfig: SpeechSDK.SpeechConfig, synthesizer:
14
14
  }
15
15
 
16
16
  export async function synthesizeSpeech(ssml: string, config: TtsConfig): Promise<ArrayBuffer> {
17
+ if (config.signal?.aborted) {
18
+ throw createSpeechSdkError("Speech synthesis was cancelled.");
19
+ }
20
+
17
21
  const speechConfig = createSpeechConfig(config);
18
22
  const synthesizer = new SpeechSDK.SpeechSynthesizer(speechConfig, null);
19
23
 
20
24
  return await new Promise<ArrayBuffer>((resolve, reject) => {
21
25
  let resourcesClosed = false;
26
+ let settled = false;
27
+ let timeout: ReturnType<typeof setTimeout> | undefined;
28
+ let abortHandler: (() => void) | undefined;
29
+ const cleanup = () => {
30
+ if (timeout) clearTimeout(timeout);
31
+ if (abortHandler) config.signal?.removeEventListener("abort", abortHandler);
32
+ };
22
33
  const closeResources = () => {
23
34
  if (resourcesClosed) return;
24
35
  resourcesClosed = true;
25
36
  closeSpeechResources(speechConfig, synthesizer);
26
37
  };
27
38
  const rejectWithError = (error: unknown) => {
39
+ if (settled) return;
40
+ settled = true;
41
+ cleanup();
28
42
  closeResources();
29
43
  reject(createSpeechSdkError(error));
30
44
  };
31
45
 
32
46
  const cb = (result: SpeechSDK.SpeechSynthesisResult) => {
47
+ if (settled) return;
33
48
  const { reason, errorDetails } = result;
34
49
  if (reason !== SpeechSDK.ResultReason.SynthesizingAudioCompleted) {
35
50
  const err = errorDetails || `Speech synthesis failed with reason ${reason}.`;
36
51
  rejectWithError(err);
37
52
  return;
38
53
  }
54
+ settled = true;
55
+ cleanup();
39
56
  closeResources();
40
57
  resolve(result.audioData);
41
58
  };
42
59
 
43
60
  try {
61
+ if (config.signal) {
62
+ abortHandler = () => rejectWithError("Speech synthesis was cancelled.");
63
+ config.signal.addEventListener("abort", abortHandler, { once: true });
64
+ }
65
+ if (config.timeoutMs !== undefined && config.timeoutMs > 0) {
66
+ timeout = setTimeout(
67
+ () => rejectWithError(`Speech synthesis timed out after ${config.timeoutMs} ms.`),
68
+ config.timeoutMs,
69
+ );
70
+ }
44
71
  synthesizer.speakSsmlAsync(ssml, cb, rejectWithError);
45
72
  } catch (error) {
46
73
  rejectWithError(error);
package/src/types.ts CHANGED
@@ -1,4 +1,6 @@
1
1
  export interface TtsConfig {
2
+ signal?: AbortSignal;
3
+ timeoutMs?: number;
2
4
  endpoint?: string;
3
5
  subscriptionKey: string;
4
6
  region: string;
@@ -13,6 +15,8 @@ export interface AzureTtsLogger {
13
15
  }
14
16
 
15
17
  export interface AzureTtsClientOptions {
18
+ signal?: AbortSignal;
19
+ timeoutMs?: number;
16
20
  subscriptionKey: string;
17
21
  region: string;
18
22
  endpoint?: string;
package/test/edge-cases.test.ts CHANGED
@@ -76,3 +76,46 @@ test("AzureTtsClient reports Speech SDK callback failures", async (t) => {
76
76
  },
77
77
  );
78
78
  });
79
+
80
+ test("synthesizeSpeech aborts a timed-out synthesis and closes resources", async (t) => {
81
+ const originalFromEndpoint = SpeechSDK.SpeechConfig.fromEndpoint;
82
+ let closeCount = 0;
83
+ t.mock.method(SpeechSDK.SpeechConfig, "fromEndpoint", (endpoint, subscriptionKey) =>
84
+ originalFromEndpoint(endpoint, String(subscriptionKey)),
85
+ );
86
+ t.mock.method(SpeechSDK.SpeechSynthesizer.prototype, "speakSsmlAsync", () => {});
87
+ t.mock.method(SpeechSDK.SpeechSynthesizer.prototype, "close", () => {
88
+ closeCount += 1;
89
+ });
90
+
91
+ await assert.rejects(
92
+ synthesizeSpeech("<speak>Hello</speak>", {
93
+ endpoint: "https://speech.example.test/cognitiveservices/v1",
94
+ subscriptionKey: "subscription-key",
95
+ region: "japaneast",
96
+ timeoutMs: 10,
97
+ }),
98
+ /timed out after 10 ms/,
99
+ );
100
+ assert.equal(closeCount, 1);
101
+ });
102
+
103
+ test("synthesizeSpeech can be cancelled with an AbortSignal", async (t) => {
104
+ const originalFromEndpoint = SpeechSDK.SpeechConfig.fromEndpoint;
105
+ const controller = new AbortController();
106
+ t.mock.method(SpeechSDK.SpeechConfig, "fromEndpoint", (endpoint, subscriptionKey) =>
107
+ originalFromEndpoint(endpoint, String(subscriptionKey)),
108
+ );
109
+ t.mock.method(SpeechSDK.SpeechSynthesizer.prototype, "speakSsmlAsync", () => {});
110
+ t.mock.method(SpeechSDK.SpeechSynthesizer.prototype, "close", () => {});
111
+
112
+ const promise = synthesizeSpeech("<speak>Hello</speak>", {
113
+ endpoint: "https://speech.example.test/cognitiveservices/v1",
114
+ subscriptionKey: "subscription-key",
115
+ region: "japaneast",
116
+ signal: controller.signal,
117
+ });
118
+ controller.abort();
119
+
120
+ await assert.rejects(promise, /cancelled/);
121
+ });
package/test/output-formats.test.ts CHANGED
@@ -12,6 +12,14 @@ test("defines the default Azure Speech output format", () => {
12
12
  });
13
13
 
14
14
  test("resolves supported output formats", () => {
15
+ assert.equal(
16
+ resolveOutputFormat("audio-16khz-32kbitrate-mono-mp3"),
17
+ SpeechSDK.SpeechSynthesisOutputFormat.Audio16Khz32KBitRateMonoMp3,
18
+ );
19
+ assert.equal(
20
+ resolveOutputFormat("audio-24khz-48kbitrate-mono-mp3"),
21
+ SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz48KBitRateMonoMp3,
22
+ );
15
23
  assert.equal(
16
24
  resolveOutputFormat("audio-24khz-160kbitrate-mono-mp3"),
17
25
  SpeechSDK.SpeechSynthesisOutputFormat.Audio24Khz160KBitRateMonoMp3,
package/test/synthesis.test.ts ADDED
@@ -0,0 +1,100 @@
1
+ import assert from "node:assert/strict";
2
+ import test, { type TestContext } from "node:test";
3
+ import * as SpeechSDK from "microsoft-cognitiveservices-speech-sdk";
4
+ import { synthesizeSpeech } from "../src/index.ts";
5
+
6
+ const endpoint = "https://speech.example.test/cognitiveservices/v1";
7
+ const subscriptionKey = "subscription-key";
8
+ const region = "japaneast";
9
+
10
+ function installHangingSynthesisMock(
11
+ testContext: TestContext,
12
+ onResult: (callback: (result: SpeechSDK.SpeechSynthesisResult) => void) => void,
13
+ onClose: () => void,
14
+ ): void {
15
+ const originalFromEndpoint = SpeechSDK.SpeechConfig.fromEndpoint;
16
+ testContext.mock.method(SpeechSDK.SpeechConfig, "fromEndpoint", (speechEndpoint, key) =>
17
+ originalFromEndpoint(speechEndpoint, String(key)),
18
+ );
19
+ testContext.mock.method(SpeechSDK.SpeechSynthesizer.prototype, "speakSsmlAsync", (_ssml, callback) => {
20
+ if (callback) onResult(callback);
21
+ });
22
+ testContext.mock.method(SpeechSDK.SpeechSynthesizer.prototype, "close", onClose);
23
+ }
24
+
25
+ test("synthesizeSpeech aborts the SDK request and settles its promise", async (t) => {
26
+ const controller = new AbortController();
27
+ let closeCount = 0;
28
+ let resultCallback: ((result: SpeechSDK.SpeechSynthesisResult) => void) | undefined;
29
+ installHangingSynthesisMock(
30
+ t,
31
+ (callback) => (resultCallback = callback),
32
+ () => (closeCount += 1),
33
+ );
34
+
35
+ const promise = synthesizeSpeech("<speak>Hello</speak>", {
36
+ endpoint,
37
+ subscriptionKey,
38
+ region,
39
+ signal: controller.signal,
40
+ });
41
+ controller.abort();
42
+
43
+ await assert.rejects(promise, /Speech synthesis was cancelled/);
44
+ assert.equal(closeCount, 1);
45
+
46
+ resultCallback?.({
47
+ audioData: new ArrayBuffer(1),
48
+ errorDetails: "",
49
+ reason: SpeechSDK.ResultReason.SynthesizingAudioCompleted,
50
+ } as SpeechSDK.SpeechSynthesisResult);
51
+ assert.equal(closeCount, 1);
52
+ });
53
+
54
+ test("synthesizeSpeech aborts the SDK request on timeout and settles its promise", async (t) => {
55
+ let closeCount = 0;
56
+ let resultCallback: ((result: SpeechSDK.SpeechSynthesisResult) => void) | undefined;
57
+ installHangingSynthesisMock(
58
+ t,
59
+ (callback) => (resultCallback = callback),
60
+ () => (closeCount += 1),
61
+ );
62
+
63
+ const promise = synthesizeSpeech("<speak>Hello</speak>", {
64
+ endpoint,
65
+ subscriptionKey,
66
+ region,
67
+ timeoutMs: 10,
68
+ });
69
+
70
+ await assert.rejects(promise, /Speech synthesis timed out after 10 ms/);
71
+ assert.equal(closeCount, 1);
72
+
73
+ resultCallback?.({
74
+ audioData: new ArrayBuffer(1),
75
+ errorDetails: "",
76
+ reason: SpeechSDK.ResultReason.SynthesizingAudioCompleted,
77
+ } as SpeechSDK.SpeechSynthesisResult);
78
+ assert.equal(closeCount, 1);
79
+ });
80
+
81
+ const azureKey = process.env.AZURE_SPEECH_KEY;
82
+ const azureRegion = process.env.AZURE_SPEECH_REGION;
83
+
84
+ test("synthesizeSpeech reaches the Azure Speech API when credentials are configured", {
85
+ skip: !azureKey || !azureRegion,
86
+ timeout: 60_000,
87
+ }, async () => {
88
+ if (!azureKey || !azureRegion) return;
89
+
90
+ const audio = await synthesizeSpeech(
91
+ '<speak version="1.0" xml:lang="en-US"><voice name="en-US-JennyNeural">Connectivity check.</voice></speak>',
92
+ {
93
+ subscriptionKey: azureKey,
94
+ region: azureRegion,
95
+ timeoutMs: 30_000,
96
+ },
97
+ );
98
+
99
+ assert.ok(audio.byteLength > 0);
100
+ });