@ssml-builder-js/azure-tts-client 2.13.0 → 2.14.0

package/CHANGELOG.md

@@ -1,5 +1,16 @@
 # @ssml-builder-js/azure-tts-client
 
+## 2.14.0
+
+### Minor Changes
+
+- Add safe container-aware audio merging, preflight-validated SSML chunk synthesis with structured progress events, source mapping metadata, controlled URL validation, and voice capability details in the Visual Editor.
+
+### Patch Changes
+
+- Updated dependencies
+  - @ssml-builder-js/ssml-core@2.14.0
+
 ## 2.13.0
 
 ### Minor Changes

package/dist/index.d.mts

@@ -1,4 +1,4 @@
-import { SsmlDiagnostic, AzureValidationOptions } from '@ssml-builder-js/ssml-core';
+import { SsmlTextRange, SsmlDiagnostic, AzureValidationOptions } from '@ssml-builder-js/ssml-core';
 
 interface TtsConfig {
     signal?: AbortSignal;
@@ -12,13 +12,13 @@ interface TtsConfig {
         start: number;
         end: number;
     };
-    /** Reports completion of a chunk when using synthesizeSsmlChunks. */
-    onProgress?: (event: {
-        currentChunk: number;
-        totalChunks: number;
-        percent: number;
-    }) => void;
+    /** Reports chunk lifecycle events when using chunk synthesis. */
+    onProgress?: (event: SynthesisProgressEvent) => void;
+    /** Metadata used to map synchronization events back to the source document. */
+    chunkIndex?: number;
+    sourceNodePath?: string[];
 }
+type SynthesisChunkStatus = "pending" | "synthesizing" | "success" | "failed";
 interface SsmlSynthesisBoundary {
     text: string;
     audioOffsetMs: number;
@@ -27,6 +27,14 @@ interface SsmlSynthesisBoundary {
         start: number;
         end: number;
     };
+    /** Chunk that produced this event. */
+    chunkIndex?: number;
+    /** Path of the source SSML node, when available. */
+    sourceNodePath?: string[];
+    /** Original text range represented by this event. */
+    originalTextRange?: SsmlTextRange;
+    /** Audio offset within the originating chunk before merge. */
+    chunkAudioOffsetMs?: number;
     requestId?: string;
 }
 interface SsmlSynthesisViseme {
@@ -36,6 +44,10 @@ interface SsmlSynthesisViseme {
         start: number;
         end: number;
     };
+    chunkIndex?: number;
+    sourceNodePath?: string[];
+    originalTextRange?: SsmlTextRange;
+    chunkAudioOffsetMs?: number;
     requestId?: string;
 }
 interface SsmlSynthesisBookmark {
@@ -45,6 +57,10 @@ interface SsmlSynthesisBookmark {
         start: number;
         end: number;
     };
+    chunkIndex?: number;
+    sourceNodePath?: string[];
+    originalTextRange?: SsmlTextRange;
+    chunkAudioOffsetMs?: number;
     requestId?: string;
 }
 /** Audio and Azure Speech synchronization events emitted for one SSML request. */
@@ -72,14 +88,21 @@ interface SsmlSynthesisChunk {
         start: number;
         end: number;
     };
+    sourceNodePath?: string[];
 }
 interface SynthesizeChunksOptions {
     onProgress?: (event: SynthesisProgressEvent) => void;
 }
 interface SynthesisProgressEvent {
+    /** 1-based completed chunk count retained for backward compatibility. */
     currentChunk: number;
     totalChunks: number;
     percent: number;
+    chunkIndex: number;
+    originalTextRange?: SsmlTextRange;
+    status: SynthesisChunkStatus;
+    durationMs: number;
+    error?: unknown;
 }
 interface AzureTtsLogger {
     debug?: (...args: unknown[]) => void;
@@ -109,12 +132,23 @@ declare class AzureTtsSdkError extends AzureTtsError {
     readonly errorDetails: string;
     constructor(errorDetails: string);
 }
+/** Thrown when audio buffers require container re-multiplexing before they can be merged. */
+declare class UnsupportedMergeFormatError extends Error {
+    readonly format: string;
+    constructor(format: string);
+}
 
 interface SsmlValidationError {
     readonly kind: "validation";
     readonly message: string;
     readonly diagnostics: readonly SsmlDiagnostic[];
 }
+declare class ChunkValidationError extends Error {
+    readonly kind: "chunk-validation";
+    readonly chunkIndex: number;
+    readonly diagnostics: readonly SsmlDiagnostic[];
+    constructor(chunkIndex: number, diagnostics: readonly SsmlDiagnostic[]);
+}
 type Result<T, E> = {
     readonly ok: true;
     readonly success: true;
@@ -141,11 +175,22 @@ interface SynthesizeSsmlSafeOptions extends AzureValidationOptions {
     /** Optional nested form for callers that want to keep validation settings grouped. */
     validation?: AzureValidationOptions;
 }
+interface SynthesizeSsmlChunksSafeOptions extends AzureValidationOptions {
+    validation?: AzureValidationOptions;
+    outputFormat?: string;
+    onProgress?: (event: SynthesisProgressEvent) => void;
+}
+type SsmlSynthesisChunksSafeResult = Result<SsmlSynthesisResult, never> | Result<never, ChunkValidationError> | Result<never, AzureTtsError>;
 interface SynthesisClient {
     synthesizeSsml(ssml: string): Promise<SsmlSynthesisResult>;
+    synthesizeChunks?(chunks: readonly (SsmlSynthesisChunk | string)[], options?: {
+        onProgress?: (event: SynthesisProgressEvent) => void;
+    }): Promise<SsmlSynthesisResult>;
 }
 /** Validates SSML before invoking Azure and converts validation/API failures to one result shape. */
 declare function synthesizeSsmlSafe(client: Pick<AzureTtsClient, "synthesizeSsml"> | SynthesisClient, ssml: string, options?: SynthesizeSsmlSafeOptions): Promise<SsmlSynthesisSafeResult>;
+/** Validates every chunk before synthesis and returns a chunk-addressable result. */
+declare function synthesizeSsmlChunksSafe(client: Pick<AzureTtsClient, "synthesizeSsml" | "synthesizeChunks"> | SynthesisClient, chunks: readonly (SsmlSynthesisChunk | string)[], options?: SynthesizeSsmlChunksSafeOptions): Promise<SsmlSynthesisChunksSafeResult>;
 
 declare class AzureTtsClient {
     #private;
@@ -154,13 +199,21 @@ declare class AzureTtsClient {
     synthesizeSsml(ssml: string): Promise<SsmlSynthesisResult>;
     synthesizeChunks(chunks: readonly (SsmlSynthesisChunk | string)[], options?: SynthesizeChunksOptions): Promise<SsmlSynthesisResult>;
     synthesizeSsmlSafe(ssml: string, options?: SynthesizeSsmlSafeOptions): Promise<SsmlSynthesisSafeResult>;
+    synthesizeChunksSafe(chunks: readonly (SsmlSynthesisChunk | string)[], options?: SynthesizeSsmlChunksSafeOptions): Promise<SsmlSynthesisChunksSafeResult>;
+    synthesizeSsmlChunksSafe(chunks: readonly (SsmlSynthesisChunk | string)[], options?: SynthesizeSsmlChunksSafeOptions): Promise<SsmlSynthesisChunksSafeResult>;
 }
 
+type MergeAudioFormat = "wav" | "mp3" | "raw";
+/** Returns whether the named output format can be safely concatenated without re-multiplexing. */
+declare function resolveMergeAudioFormat(format: string): MergeAudioFormat | undefined;
+declare function canMergeAudioFormat(format: string): boolean;
+/** Merges audio buffers while preserving the invariants of supported containers. */
+declare function mergeAudioBuffers(buffers: readonly ArrayBuffer[], format: string): ArrayBuffer;
 declare function synthesizeSsml(ssml: string, config: TtsConfig): Promise<SsmlSynthesisResult>;
 /** Synthesizes chunks sequentially, annotates synchronization events, and merges the results. */
 declare function synthesizeSsmlChunks(chunks: readonly (SsmlSynthesisChunk | string)[], config: TtsConfig): Promise<SsmlSynthesisResult>;
 /** Concatenates audio buffers and shifts all synchronization events by prior chunk durations. */
-declare function mergeSynthesisResults(results: readonly SsmlSynthesisResult[]): SsmlSynthesisResult;
+declare function mergeSynthesisResults(results: readonly SsmlSynthesisResult[], format?: string): SsmlSynthesisResult;
 /** Backward-compatible audio-only synthesis helper. */
 declare function synthesizeSpeech(ssml: string, config: TtsConfig): Promise<ArrayBuffer>;
 
@@ -173,6 +226,9 @@ interface AzureVoiceCatalogVoice {
     locale: string;
     secondaryLocales?: readonly string[];
     styles?: readonly string[];
+    supportedTags?: readonly string[];
+    unsupportedTags?: readonly string[];
+    models?: readonly string[];
     regions: readonly string[];
     status?: "ga" | "preview" | "deprecated";
 }
@@ -189,4 +245,4 @@ interface AzureVoiceCatalog {
 /** Fetches and deduplicates the current Azure Speech voice catalog for one or more regions. */
 declare function fetchAzureVoiceCatalog(options: FetchAzureVoiceCatalogOptions): Promise<AzureVoiceCatalog>;
 
-export { type AzureApiErrorResult, type SsmlValidationError as AzureSsmlValidationError, AzureTtsClient, type AzureTtsClientOptions, AzureTtsError, type AzureTtsLogger, AzureTtsSdkError, type AzureVoiceCatalog, type AzureVoiceCatalogVoice, type FetchAzureVoiceCatalogOptions, type FetchedAzureVoiceCatalogMetadata, type Result, type SsmlSynthesisBookmark, type SsmlSynthesisBoundary, type SsmlSynthesisChunk, type SsmlSynthesisResult, type SsmlSynthesisSafeResult, type SsmlSynthesisViseme, type Success, type SynthesisProgressEvent, type SynthesisResult, type SynthesizeChunksOptions, type SynthesizeSsmlSafeOptions, type TtsConfig, type ValidationErrorResult, fetchAzureVoiceCatalog, mergeSynthesisResults, synthesizeSpeech, synthesizeSsml, synthesizeSsmlChunks, synthesizeSsmlSafe };
+export { type AzureApiErrorResult, type SsmlValidationError as AzureSsmlValidationError, AzureTtsClient, type AzureTtsClientOptions, AzureTtsError, type AzureTtsLogger, AzureTtsSdkError, type AzureVoiceCatalog, type AzureVoiceCatalogVoice, ChunkValidationError, type FetchAzureVoiceCatalogOptions, type FetchedAzureVoiceCatalogMetadata, type MergeAudioFormat, type Result, type SsmlSynthesisBookmark, type SsmlSynthesisBoundary, type SsmlSynthesisChunk, type SsmlSynthesisChunksSafeResult, type SsmlSynthesisResult, type SsmlSynthesisSafeResult, type SsmlSynthesisViseme, type Success, type SynthesisChunkStatus, type SynthesisProgressEvent, type SynthesisResult, type SynthesizeChunksOptions, type SynthesizeSsmlChunksSafeOptions, type SynthesizeSsmlSafeOptions, type TtsConfig, UnsupportedMergeFormatError, type ValidationErrorResult, canMergeAudioFormat, fetchAzureVoiceCatalog, mergeAudioBuffers, mergeSynthesisResults, resolveMergeAudioFormat, synthesizeSpeech, synthesizeSsml, synthesizeSsmlChunks, synthesizeSsmlChunksSafe, synthesizeSsmlSafe };

package/dist/index.d.ts

@@ -1,4 +1,4 @@
-import { SsmlDiagnostic, AzureValidationOptions } from '@ssml-builder-js/ssml-core';
+import { SsmlTextRange, SsmlDiagnostic, AzureValidationOptions } from '@ssml-builder-js/ssml-core';
 
 interface TtsConfig {
     signal?: AbortSignal;
@@ -12,13 +12,13 @@ interface TtsConfig {
         start: number;
         end: number;
     };
-    /** Reports completion of a chunk when using synthesizeSsmlChunks. */
-    onProgress?: (event: {
-        currentChunk: number;
-        totalChunks: number;
-        percent: number;
-    }) => void;
+    /** Reports chunk lifecycle events when using chunk synthesis. */
+    onProgress?: (event: SynthesisProgressEvent) => void;
+    /** Metadata used to map synchronization events back to the source document. */
+    chunkIndex?: number;
+    sourceNodePath?: string[];
 }
+type SynthesisChunkStatus = "pending" | "synthesizing" | "success" | "failed";
 interface SsmlSynthesisBoundary {
     text: string;
     audioOffsetMs: number;
@@ -27,6 +27,14 @@ interface SsmlSynthesisBoundary {
         start: number;
         end: number;
     };
+    /** Chunk that produced this event. */
+    chunkIndex?: number;
+    /** Path of the source SSML node, when available. */
+    sourceNodePath?: string[];
+    /** Original text range represented by this event. */
+    originalTextRange?: SsmlTextRange;
+    /** Audio offset within the originating chunk before merge. */
+    chunkAudioOffsetMs?: number;
     requestId?: string;
 }
 interface SsmlSynthesisViseme {
@@ -36,6 +44,10 @@ interface SsmlSynthesisViseme {
         start: number;
         end: number;
     };
+    chunkIndex?: number;
+    sourceNodePath?: string[];
+    originalTextRange?: SsmlTextRange;
+    chunkAudioOffsetMs?: number;
     requestId?: string;
 }
 interface SsmlSynthesisBookmark {
@@ -45,6 +57,10 @@ interface SsmlSynthesisBookmark {
         start: number;
         end: number;
     };
+    chunkIndex?: number;
+    sourceNodePath?: string[];
+    originalTextRange?: SsmlTextRange;
+    chunkAudioOffsetMs?: number;
     requestId?: string;
 }
 /** Audio and Azure Speech synchronization events emitted for one SSML request. */
@@ -72,14 +88,21 @@ interface SsmlSynthesisChunk {
         start: number;
         end: number;
     };
+    sourceNodePath?: string[];
 }
 interface SynthesizeChunksOptions {
     onProgress?: (event: SynthesisProgressEvent) => void;
 }
 interface SynthesisProgressEvent {
+    /** 1-based completed chunk count retained for backward compatibility. */
     currentChunk: number;
     totalChunks: number;
     percent: number;
+    chunkIndex: number;
+    originalTextRange?: SsmlTextRange;
+    status: SynthesisChunkStatus;
+    durationMs: number;
+    error?: unknown;
 }
 interface AzureTtsLogger {
     debug?: (...args: unknown[]) => void;
@@ -109,12 +132,23 @@ declare class AzureTtsSdkError extends AzureTtsError {
     readonly errorDetails: string;
     constructor(errorDetails: string);
 }
+/** Thrown when audio buffers require container re-multiplexing before they can be merged. */
+declare class UnsupportedMergeFormatError extends Error {
+    readonly format: string;
+    constructor(format: string);
+}
 
 interface SsmlValidationError {
     readonly kind: "validation";
     readonly message: string;
     readonly diagnostics: readonly SsmlDiagnostic[];
 }
+declare class ChunkValidationError extends Error {
+    readonly kind: "chunk-validation";
+    readonly chunkIndex: number;
+    readonly diagnostics: readonly SsmlDiagnostic[];
+    constructor(chunkIndex: number, diagnostics: readonly SsmlDiagnostic[]);
+}
 type Result<T, E> = {
     readonly ok: true;
     readonly success: true;
@@ -141,11 +175,22 @@ interface SynthesizeSsmlSafeOptions extends AzureValidationOptions {
     /** Optional nested form for callers that want to keep validation settings grouped. */
     validation?: AzureValidationOptions;
 }
+interface SynthesizeSsmlChunksSafeOptions extends AzureValidationOptions {
+    validation?: AzureValidationOptions;
+    outputFormat?: string;
+    onProgress?: (event: SynthesisProgressEvent) => void;
+}
+type SsmlSynthesisChunksSafeResult = Result<SsmlSynthesisResult, never> | Result<never, ChunkValidationError> | Result<never, AzureTtsError>;
 interface SynthesisClient {
     synthesizeSsml(ssml: string): Promise<SsmlSynthesisResult>;
+    synthesizeChunks?(chunks: readonly (SsmlSynthesisChunk | string)[], options?: {
+        onProgress?: (event: SynthesisProgressEvent) => void;
+    }): Promise<SsmlSynthesisResult>;
 }
 /** Validates SSML before invoking Azure and converts validation/API failures to one result shape. */
 declare function synthesizeSsmlSafe(client: Pick<AzureTtsClient, "synthesizeSsml"> | SynthesisClient, ssml: string, options?: SynthesizeSsmlSafeOptions): Promise<SsmlSynthesisSafeResult>;
+/** Validates every chunk before synthesis and returns a chunk-addressable result. */
+declare function synthesizeSsmlChunksSafe(client: Pick<AzureTtsClient, "synthesizeSsml" | "synthesizeChunks"> | SynthesisClient, chunks: readonly (SsmlSynthesisChunk | string)[], options?: SynthesizeSsmlChunksSafeOptions): Promise<SsmlSynthesisChunksSafeResult>;
 
 declare class AzureTtsClient {
     #private;
@@ -154,13 +199,21 @@ declare class AzureTtsClient {
     synthesizeSsml(ssml: string): Promise<SsmlSynthesisResult>;
     synthesizeChunks(chunks: readonly (SsmlSynthesisChunk | string)[], options?: SynthesizeChunksOptions): Promise<SsmlSynthesisResult>;
     synthesizeSsmlSafe(ssml: string, options?: SynthesizeSsmlSafeOptions): Promise<SsmlSynthesisSafeResult>;
+    synthesizeChunksSafe(chunks: readonly (SsmlSynthesisChunk | string)[], options?: SynthesizeSsmlChunksSafeOptions): Promise<SsmlSynthesisChunksSafeResult>;
+    synthesizeSsmlChunksSafe(chunks: readonly (SsmlSynthesisChunk | string)[], options?: SynthesizeSsmlChunksSafeOptions): Promise<SsmlSynthesisChunksSafeResult>;
 }
 
+type MergeAudioFormat = "wav" | "mp3" | "raw";
+/** Returns whether the named output format can be safely concatenated without re-multiplexing. */
+declare function resolveMergeAudioFormat(format: string): MergeAudioFormat | undefined;
+declare function canMergeAudioFormat(format: string): boolean;
+/** Merges audio buffers while preserving the invariants of supported containers. */
+declare function mergeAudioBuffers(buffers: readonly ArrayBuffer[], format: string): ArrayBuffer;
 declare function synthesizeSsml(ssml: string, config: TtsConfig): Promise<SsmlSynthesisResult>;
 /** Synthesizes chunks sequentially, annotates synchronization events, and merges the results. */
 declare function synthesizeSsmlChunks(chunks: readonly (SsmlSynthesisChunk | string)[], config: TtsConfig): Promise<SsmlSynthesisResult>;
 /** Concatenates audio buffers and shifts all synchronization events by prior chunk durations. */
-declare function mergeSynthesisResults(results: readonly SsmlSynthesisResult[]): SsmlSynthesisResult;
+declare function mergeSynthesisResults(results: readonly SsmlSynthesisResult[], format?: string): SsmlSynthesisResult;
 /** Backward-compatible audio-only synthesis helper. */
 declare function synthesizeSpeech(ssml: string, config: TtsConfig): Promise<ArrayBuffer>;
 
@@ -173,6 +226,9 @@ interface AzureVoiceCatalogVoice {
     locale: string;
     secondaryLocales?: readonly string[];
     styles?: readonly string[];
+    supportedTags?: readonly string[];
+    unsupportedTags?: readonly string[];
+    models?: readonly string[];
     regions: readonly string[];
     status?: "ga" | "preview" | "deprecated";
 }
@@ -189,4 +245,4 @@ interface AzureVoiceCatalog {
 /** Fetches and deduplicates the current Azure Speech voice catalog for one or more regions. */
 declare function fetchAzureVoiceCatalog(options: FetchAzureVoiceCatalogOptions): Promise<AzureVoiceCatalog>;
 
-export { type AzureApiErrorResult, type SsmlValidationError as AzureSsmlValidationError, AzureTtsClient, type AzureTtsClientOptions, AzureTtsError, type AzureTtsLogger, AzureTtsSdkError, type AzureVoiceCatalog, type AzureVoiceCatalogVoice, type FetchAzureVoiceCatalogOptions, type FetchedAzureVoiceCatalogMetadata, type Result, type SsmlSynthesisBookmark, type SsmlSynthesisBoundary, type SsmlSynthesisChunk, type SsmlSynthesisResult, type SsmlSynthesisSafeResult, type SsmlSynthesisViseme, type Success, type SynthesisProgressEvent, type SynthesisResult, type SynthesizeChunksOptions, type SynthesizeSsmlSafeOptions, type TtsConfig, type ValidationErrorResult, fetchAzureVoiceCatalog, mergeSynthesisResults, synthesizeSpeech, synthesizeSsml, synthesizeSsmlChunks, synthesizeSsmlSafe };
+export { type AzureApiErrorResult, type SsmlValidationError as AzureSsmlValidationError, AzureTtsClient, type AzureTtsClientOptions, AzureTtsError, type AzureTtsLogger, AzureTtsSdkError, type AzureVoiceCatalog, type AzureVoiceCatalogVoice, ChunkValidationError, type FetchAzureVoiceCatalogOptions, type FetchedAzureVoiceCatalogMetadata, type MergeAudioFormat, type Result, type SsmlSynthesisBookmark, type SsmlSynthesisBoundary, type SsmlSynthesisChunk, type SsmlSynthesisChunksSafeResult, type SsmlSynthesisResult, type SsmlSynthesisSafeResult, type SsmlSynthesisViseme, type Success, type SynthesisChunkStatus, type SynthesisProgressEvent, type SynthesisResult, type SynthesizeChunksOptions, type SynthesizeSsmlChunksSafeOptions, type SynthesizeSsmlSafeOptions, type TtsConfig, UnsupportedMergeFormatError, type ValidationErrorResult, canMergeAudioFormat, fetchAzureVoiceCatalog, mergeAudioBuffers, mergeSynthesisResults, resolveMergeAudioFormat, synthesizeSpeech, synthesizeSsml, synthesizeSsmlChunks, synthesizeSsmlChunksSafe, synthesizeSsmlSafe };