@mastra/voice-deepgram 0.0.0-revert-schema-20250416221206 → 0.0.0-safe-stringify-telemetry-20251205024938
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.
- package/CHANGELOG.md +1721 -0
- package/LICENSE.md +11 -42
- package/README.md +10 -7
- package/dist/index.cjs +64 -15
- package/dist/index.cjs.map +1 -0
- package/dist/index.d.ts +52 -4
- package/dist/index.d.ts.map +1 -0
- package/dist/index.js +64 -15
- package/dist/index.js.map +1 -0
- package/dist/voices.d.ts +13 -0
- package/dist/voices.d.ts.map +1 -0
- package/package.json +37 -16
- package/dist/_tsup-dts-rollup.d.cts +0 -50
- package/dist/_tsup-dts-rollup.d.ts +0 -50
- package/dist/index.d.cts +0 -4
package/LICENSE.md
CHANGED
|
@@ -1,46 +1,15 @@
|
|
|
1
|
-
#
|
|
1
|
+
# Apache License 2.0
|
|
2
2
|
|
|
3
|
-
Copyright (c) 2025
|
|
3
|
+
Copyright (c) 2025 Kepler Software, Inc.
|
|
4
4
|
|
|
5
|
-
|
|
6
|
-
|
|
5
|
+
Licensed under the Apache License, Version 2.0 (the "License");
|
|
6
|
+
you may not use this file except in compliance with the License.
|
|
7
|
+
You may obtain a copy of the License at
|
|
7
8
|
|
|
8
|
-
|
|
9
|
-
The licensor grants you a non-exclusive, royalty-free, worldwide, non-sublicensable, non-transferable license to use, copy, distribute, make available, and prepare derivative works of the software, in each case subject to the limitations and conditions below
|
|
9
|
+
http://www.apache.org/licenses/LICENSE-2.0
|
|
10
10
|
|
|
11
|
-
|
|
12
|
-
|
|
13
|
-
|
|
14
|
-
|
|
15
|
-
|
|
16
|
-
You may not alter, remove, or obscure any licensing, copyright, or other notices of the licensor in the software. Any use of the licensor’s trademarks is subject to applicable law.
|
|
17
|
-
|
|
18
|
-
**Patents**
|
|
19
|
-
The licensor grants you a license, under any patent claims the licensor can license, or becomes able to license, to make, have made, use, sell, offer for sale, import and have imported the software, in each case subject to the limitations and conditions in this license. This license does not cover any patent claims that you cause to be infringed by modifications or additions to the software. If you or your company make any written claim that the software infringes or contributes to infringement of any patent, your patent license for the software granted under these terms ends immediately. If your company makes such a claim, your patent license ends immediately for work on behalf of your company.
|
|
20
|
-
|
|
21
|
-
**Notices**
|
|
22
|
-
You must ensure that anyone who gets a copy of any part of the software from you also gets a copy of these terms.
|
|
23
|
-
|
|
24
|
-
If you modify the software, you must include in any modified copies of the software prominent notices stating that you have modified the software.
|
|
25
|
-
|
|
26
|
-
**No Other Rights**
|
|
27
|
-
These terms do not imply any licenses other than those expressly granted in these terms.
|
|
28
|
-
|
|
29
|
-
**Termination**
|
|
30
|
-
If you use the software in violation of these terms, such use is not licensed, and your licenses will automatically terminate. If the licensor provides you with a notice of your violation, and you cease all violation of this license no later than 30 days after you receive that notice, your licenses will be reinstated retroactively. However, if you violate these terms after such reinstatement, any additional violation of these terms will cause your licenses to terminate automatically and permanently.
|
|
31
|
-
|
|
32
|
-
**No Liability**
|
|
33
|
-
As far as the law allows, the software comes as is, without any warranty or condition, and the licensor will not be liable to you for any damages arising out of these terms or the use or nature of the software, under any kind of legal claim.
|
|
34
|
-
|
|
35
|
-
**Definitions**
|
|
36
|
-
The _licensor_ is the entity offering these terms, and the _software_ is the software the licensor makes available under these terms, including any portion of it.
|
|
37
|
-
|
|
38
|
-
_you_ refers to the individual or entity agreeing to these terms.
|
|
39
|
-
|
|
40
|
-
_your company_ is any legal entity, sole proprietorship, or other kind of organization that you work for, plus all organizations that have control over, are under the control of, or are under common control with that organization. _control_ means ownership of substantially all the assets of an entity, or the power to direct its management and policies by vote, contract, or otherwise. Control can be direct or indirect.
|
|
41
|
-
|
|
42
|
-
_your licenses_ are all the licenses granted to you for the software under these terms.
|
|
43
|
-
|
|
44
|
-
_use_ means anything you do with the software requiring one of your licenses.
|
|
45
|
-
|
|
46
|
-
_trademark_ means trademarks, service marks, and similar rights.
|
|
11
|
+
Unless required by applicable law or agreed to in writing, software
|
|
12
|
+
distributed under the License is distributed on an "AS IS" BASIS,
|
|
13
|
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
14
|
+
See the License for the specific language governing permissions and
|
|
15
|
+
limitations under the License.
|
package/README.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
# @mastra/voice-deepgram
|
|
2
2
|
|
|
3
|
-
Deepgram voice integration for Mastra, providing both Text-to-Speech (TTS) and Speech-to-Text (STT)
|
|
3
|
+
Deepgram voice integration for Mastra, providing both Text-to-Speech (TTS) and Speech-to-Text (STT) using Deepgram's Aura (TTS) and Nova (STT) families.
|
|
4
4
|
|
|
5
5
|
## Installation
|
|
6
6
|
|
|
@@ -24,14 +24,14 @@ import { DeepgramVoice } from '@mastra/voice-deepgram';
|
|
|
24
24
|
// Create voice with both speech and listening capabilities
|
|
25
25
|
const voice = new DeepgramVoice({
|
|
26
26
|
speechModel: {
|
|
27
|
-
name: 'aura
|
|
27
|
+
name: 'aura', // TTS family
|
|
28
28
|
apiKey: 'your-api-key', // Optional, can use DEEPGRAM_API_KEY env var
|
|
29
29
|
},
|
|
30
30
|
listeningModel: {
|
|
31
|
-
name: 'nova', //
|
|
31
|
+
name: 'nova', // STT family
|
|
32
32
|
apiKey: 'your-api-key', // Optional, can use DEEPGRAM_API_KEY env var
|
|
33
33
|
},
|
|
34
|
-
speaker: '
|
|
34
|
+
speaker: 'asteria-en', // default voiceId (see voice.ts)
|
|
35
35
|
});
|
|
36
36
|
|
|
37
37
|
// List available voices
|
|
@@ -39,12 +39,15 @@ const voices = await voice.getSpeakers();
|
|
|
39
39
|
|
|
40
40
|
// Generate speech
|
|
41
41
|
const audioStream = await voice.speak('Hello from Mastra!', {
|
|
42
|
-
speaker: '
|
|
43
|
-
speed: 1.0, // Optional: adjust speech speed
|
|
42
|
+
speaker: 'hera-en', // override speaker voice
|
|
44
43
|
});
|
|
45
44
|
|
|
46
45
|
// Convert speech to text
|
|
47
|
-
const
|
|
46
|
+
const result = await voice.listen(audioStream, {
|
|
47
|
+
diarize: true,
|
|
48
|
+
diarize_speaker_count: 2,
|
|
49
|
+
});
|
|
50
|
+
console.log(result.transcript);
|
|
48
51
|
```
|
|
49
52
|
|
|
50
53
|
## Features
|
package/dist/index.cjs
CHANGED
|
@@ -26,6 +26,9 @@ var DEEPGRAM_VOICES = [
|
|
|
26
26
|
var DeepgramVoice = class extends voice.MastraVoice {
|
|
27
27
|
speechClient;
|
|
28
28
|
listeningClient;
|
|
29
|
+
storedSpeechModel;
|
|
30
|
+
storedListeningModel;
|
|
31
|
+
storedSpeaker;
|
|
29
32
|
constructor({
|
|
30
33
|
speechModel,
|
|
31
34
|
listeningModel,
|
|
@@ -51,6 +54,14 @@ var DeepgramVoice = class extends voice.MastraVoice {
|
|
|
51
54
|
},
|
|
52
55
|
speaker
|
|
53
56
|
});
|
|
57
|
+
this.storedSpeechModel = {
|
|
58
|
+
name: speechModel?.name ?? defaultSpeechModel.name,
|
|
59
|
+
apiKey: speechModel?.apiKey ?? defaultSpeechModel.apiKey
|
|
60
|
+
};
|
|
61
|
+
this.storedListeningModel = {
|
|
62
|
+
name: listeningModel?.name ?? defaultListeningModel.name,
|
|
63
|
+
apiKey: listeningModel?.apiKey ?? defaultListeningModel.apiKey
|
|
64
|
+
};
|
|
54
65
|
const speechApiKey = speechModel?.apiKey || defaultApiKey;
|
|
55
66
|
const listeningApiKey = listeningModel?.apiKey || defaultApiKey;
|
|
56
67
|
if (!speechApiKey && !listeningApiKey) {
|
|
@@ -62,7 +73,7 @@ var DeepgramVoice = class extends voice.MastraVoice {
|
|
|
62
73
|
if (listeningApiKey) {
|
|
63
74
|
this.listeningClient = sdk.createClient(listeningApiKey);
|
|
64
75
|
}
|
|
65
|
-
this.
|
|
76
|
+
this.storedSpeaker = speaker || "asteria-en";
|
|
66
77
|
}
|
|
67
78
|
async getSpeakers() {
|
|
68
79
|
return this.traced(async () => {
|
|
@@ -96,18 +107,15 @@ var DeepgramVoice = class extends voice.MastraVoice {
|
|
|
96
107
|
if (!this.speechClient) {
|
|
97
108
|
throw new Error("No speech client configured");
|
|
98
109
|
}
|
|
99
|
-
|
|
100
|
-
|
|
101
|
-
|
|
102
|
-
} else if (this.speaker) {
|
|
103
|
-
model = this.speechModel?.name + "-" + this.speaker;
|
|
104
|
-
}
|
|
110
|
+
const baseModel = this.storedSpeechModel?.name;
|
|
111
|
+
const speakerId = options?.speaker || this.storedSpeaker;
|
|
112
|
+
const modelName = baseModel && speakerId ? speakerId.startsWith(`${baseModel}-`) ? speakerId : `${baseModel}-${speakerId}` : baseModel || speakerId;
|
|
105
113
|
const speakClient = this.speechClient.speak;
|
|
106
114
|
const response = await speakClient.request(
|
|
107
115
|
{ text },
|
|
108
116
|
{
|
|
109
|
-
model,
|
|
110
|
-
...options
|
|
117
|
+
model: modelName,
|
|
118
|
+
...Object.fromEntries(Object.entries(options ?? {}).filter(([k]) => k !== "speaker"))
|
|
111
119
|
}
|
|
112
120
|
);
|
|
113
121
|
const webStream = await response.getStream();
|
|
@@ -135,6 +143,25 @@ var DeepgramVoice = class extends voice.MastraVoice {
|
|
|
135
143
|
return nodeStream;
|
|
136
144
|
}, "voice.deepgram.speak")();
|
|
137
145
|
}
|
|
146
|
+
/**
|
|
147
|
+
* Checks if listening capabilities are enabled.
|
|
148
|
+
*
|
|
149
|
+
* @returns {Promise<{ enabled: boolean }>}
|
|
150
|
+
*/
|
|
151
|
+
async getListener() {
|
|
152
|
+
return { enabled: true };
|
|
153
|
+
}
|
|
154
|
+
/**
|
|
155
|
+
* Transcribes audio with optional speaker diarization.
|
|
156
|
+
*
|
|
157
|
+
* @param audioStream - Audio input stream
|
|
158
|
+
* @param options - Transcription options (diarize, language, etc.)
|
|
159
|
+
* @returns Promise resolving to:
|
|
160
|
+
* - transcript: Full transcript string
|
|
161
|
+
* - words: Array of word objects with timing and confidence
|
|
162
|
+
* - raw: Complete Deepgram API response
|
|
163
|
+
* - speakerSegments: (when diarize=true) Array of {word, speaker, start, end}
|
|
164
|
+
*/
|
|
138
165
|
async listen(audioStream, options) {
|
|
139
166
|
if (!this.listeningClient) {
|
|
140
167
|
throw new Error("Deepgram listening client not configured");
|
|
@@ -152,20 +179,42 @@ var DeepgramVoice = class extends voice.MastraVoice {
|
|
|
152
179
|
if (!this.listeningClient) {
|
|
153
180
|
throw new Error("No listening client configured");
|
|
154
181
|
}
|
|
182
|
+
const { diarize, diarize_speaker_count: _, ...restOptions } = options ?? {};
|
|
155
183
|
const { result, error } = await this.listeningClient.listen.prerecorded.transcribeFile(buffer, {
|
|
156
|
-
|
|
157
|
-
|
|
184
|
+
...restOptions,
|
|
185
|
+
model: this.storedListeningModel?.name,
|
|
186
|
+
diarize
|
|
158
187
|
});
|
|
159
188
|
if (error) {
|
|
160
189
|
throw error;
|
|
161
190
|
}
|
|
162
|
-
const
|
|
163
|
-
|
|
164
|
-
|
|
191
|
+
const channel = result.results?.channels?.[0];
|
|
192
|
+
const alt = channel?.alternatives?.[0];
|
|
193
|
+
if (!alt) {
|
|
194
|
+
return {
|
|
195
|
+
transcript: "",
|
|
196
|
+
words: [],
|
|
197
|
+
raw: result
|
|
198
|
+
};
|
|
199
|
+
}
|
|
200
|
+
const response = {
|
|
201
|
+
transcript: alt.transcript,
|
|
202
|
+
words: alt.words,
|
|
203
|
+
raw: result
|
|
204
|
+
};
|
|
205
|
+
if (diarize && alt.words) {
|
|
206
|
+
response.speakerSegments = alt.words.map((w) => ({
|
|
207
|
+
word: w.word,
|
|
208
|
+
speaker: w.speaker,
|
|
209
|
+
start: w.start,
|
|
210
|
+
end: w.end
|
|
211
|
+
}));
|
|
165
212
|
}
|
|
166
|
-
return
|
|
213
|
+
return response;
|
|
167
214
|
}, "voice.deepgram.listen")();
|
|
168
215
|
}
|
|
169
216
|
};
|
|
170
217
|
|
|
171
218
|
exports.DeepgramVoice = DeepgramVoice;
|
|
219
|
+
//# sourceMappingURL=index.cjs.map
|
|
220
|
+
//# sourceMappingURL=index.cjs.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"sources":["../src/voices.ts","../src/index.ts"],"names":["MastraVoice","createClient","PassThrough"],"mappings":";;;;;;;;;AAKO,IAAM,eAAA,GAAkB;AAAA,EAC7B,YAAA;AAAA,EACA,SAAA;AAAA,EACA,WAAA;AAAA,EACA,WAAA;AAAA,EACA,SAAA;AAAA,EACA,UAAA;AAAA,EACA,UAAA;AAAA,EACA,YAAA;AAAA,EACA,UAAA;AAAA,EACA,YAAA;AAAA,EACA,WAAA;AAAA,EACA;AACF,CAAA;;;ACKO,IAAM,aAAA,GAAN,cAA4BA,iBAAA,CAAY;AAAA,EACrC,YAAA;AAAA,EACA,eAAA;AAAA,EACA,iBAAA;AAAA,EACA,oBAAA;AAAA,EACA,aAAA;AAAA,EAER,WAAA,CAAY;AAAA,IACV,WAAA;AAAA,IACA,cAAA;AAAA,IACA;AAAA,GACF,GAA4G,EAAC,EAAG;AAC9G,IAAA,MAAM,aAAA,GAAgB,QAAQ,GAAA,CAAI,gBAAA;AAElC,IAAA,MAAM,kBAAA,GAA+D;AAAA,MACnE,IAAA,EAAM,MAAA;AAAA,MACN,MAAA,EAAQ;AAAA,KACV;AAEA,IAAA,MAAM,qBAAA,GAAkE;AAAA,MACtE,IAAA,EAAM,MAAA;AAAA,MACN,MAAA,EAAQ;AAAA,KACV;AAEA,IAAA,KAAA,CAAM;AAAA,MACJ,WAAA,EAAa;AAAA,QACX,IAAA,EAAM,WAAA,EAAa,IAAA,IAAQ,kBAAA,CAAmB,IAAA;AAAA,QAC9C,MAAA,EAAQ,WAAA,EAAa,MAAA,IAAU,kBAAA,CAAmB;AAAA,OACpD;AAAA,MACA,cAAA,EAAgB;AAAA,QACd,IAAA,EAAM,cAAA,EAAgB,IAAA,IAAQ,qBAAA,CAAsB,IAAA;AAAA,QACpD,MAAA,EAAQ,cAAA,EAAgB,MAAA,IAAU,qBAAA,CAAsB;AAAA,OAC1D;AAAA,MACA;AAAA,KACD,CAAA;AAED,IAAA,IAAA,CAAK,iBAAA,GAAoB;AAAA,MACvB,IAAA,EAAM,WAAA,EAAa,IAAA,IAAQ,kBAAA,CAAmB,IAAA;AAAA,MAC9C,MAAA,EAAQ,WAAA,EAAa,MAAA,IAAU,kBAAA,CAAmB;AAAA,KACpD;AACA,IAAA,IAAA,CAAK,oBAAA,GAAuB;AAAA,MAC1B,IAAA,EAAM,cAAA,EAAgB,IAAA,IAAQ,qBAAA,CAAsB,IAAA;AAAA,MACpD,MAAA,EAAQ,cAAA,EAAgB,MAAA,IAAU,qBAAA,CAAsB;AAAA,KAC1D;AAEA,IAAA,MAAM,YAAA,GAAe,aAAa,MAAA,IAAU,aAAA;AAC5C,IAAA,MAAM,eAAA,GAAkB,gBAAgB,MAAA,IAAU,aAAA;AAElD,IAAA,IAAI,CAAC,YAAA,IAAgB,CAAC,eAAA,EAAiB;AACrC,MAAA,MAAM,IAAI,MAAM,4FAA4F,CAAA;AAAA,IAC9G;AAEA,IAAA,IAAI,YAAA,EAAc;AAChB,MAAA,IAAA,CAAK,YAAA,GAAeC,iBAAa,YAAY,CAAA;AAAA,IAC/C;AACA,IAAA,IAAI,eAAA,EAAiB;AACnB,MAAA,IAAA,CAAK,eAAA,GAAkBA,iBAAa,eAAe,CAAA;AAAA,IACrD;AAEA,IAAA,IAAA,CAAK,gBAAgB,OAAA,IAAW,YAAA;AAAA,EAClC;AAAA,EAEA,MAAM,WAAA,GAAc;AAClB,IAAA,OAAO,IAAA,CAAK,OAAO,YAAY;AAC7B,MAAA,OAAO,eAAA,CAAgB,IAAI,CAAA,KAAA,MAAU;AAAA,QACnC,OAAA,EAAS;AAAA,OACX,CAAE,CAAA;AAAA,IACJ,CAAA,EAAG,4BAA4B,CAAA,EAAE;AAAA,EACnC;AAAA,EAEA,MAAM,KAAA,CACJ,KAAA,EACA,OAAA,EAIgC;AAChC,IAAA,IAAI,CAAC,KAAK,YAAA,EAAc;AACtB,MAAA,MAAM,IAAI,MAAM,uCAAuC,CAAA;AAAA,IACzD;AAEA,IAAA,IAAI,IAAA;AACJ,IAAA,IAAI,OAAO,UAAU,QAAA,EAAU;AAC7B,MAAA,MAAM,SAAmB,EAAC;AAC1B,MAAA,WAAA,MAAiB,SAAS,KAAA,EAAO;AAC/B,QAAA,IAAI,OAAO,UAAU,QAAA,EAAU;AAC7B,UAAA,MAAA,CAAO,IAAA,CAAK,MAAA,CAAO,IAAA,CAAK,KAAK,CAAC,CAAA;AAAA,QAChC,CAAA,MAAO;AACL,UAAA,MAAA,CAAO,KAAK,KAAK,CAAA;AAAA,QACnB;AAAA,MACF;AACA,MAAA,IAAA,GAAO,MAAA,CAAO,MAAA,CAAO,MAAM,CAAA,CAAE,SAAS,OAAO,CAAA;AAAA,IAC/C,CAAA,MAAO;AACL,MAAA,IAAA,GAAO,KAAA;AAAA,IACT;AAEA,IAAA,IAAI,IAAA,CAAK,IAAA,EAAK,CAAE,MAAA,KAAW,CAAA,EAAG;AAC5B,MAAA,MAAM,IAAI,MAAM,qBAAqB,CAAA;AAAA,IACvC;AAEA,IAAA,OAAO,IAAA,CAAK,OAAO,YAAY;AAC7B,MAAA,IAAI,CAAC,KAAK,YAAA,EAAc;AACtB,QAAA,MAAM,IAAI,MAAM,6BAA6B,CAAA;AAAA,MAC/C;AAEA,MAAA,MAAM,SAAA,GAAY,KAAK,iBAAA,EAAmB,IAAA;AAC1C,MAAA,MAAM,SAAA,GAAY,OAAA,EAAS,OAAA,IAAW,IAAA,CAAK,aAAA;AAE3C,MAAA,MAAM,SAAA,GACJ,SAAA,IAAa,SAAA,GACT,SAAA,CAAU,WAAW,CAAA,EAAG,SAAS,CAAA,CAAA,CAAG,CAAA,GAClC,YACA,CAAA,EAAG,SAAS,CAAA,CAAA,EAAI,SAAS,KAC3B,SAAA,IAAa,SAAA;AAEnB,MAAA,MAAM,WAAA,GAAc,KAAK,YAAA,CAAa,KAAA;AACtC,MAAA,MAAM,QAAA,GAAW,MAAM,WAAA,CAAY,OAAA;AAAA,QACjC,EAAE,IAAA,EAAK;AAAA,QACP;AAAA,UACE,KAAA,EAAO,SAAA;AAAA,UACP,GAAG,MAAA,CAAO,WAAA,CAAY,MAAA,CAAO,OAAA,CAAQ,WAAW,EAAE,CAAA,CAAE,MAAA,CAAO,CAAC,CAAC,CAAC,CAAA,KAAM,CAAA,KAAM,SAAS,CAAC;AAAA;AACtF,OACF;AAEA,MAAA,MAAM,SAAA,GAAY,MAAM,QAAA,CAAS,SAAA,EAAU;AAC3C,MAAA,IAAI,CAAC,SAAA,EAAW;AACd,QAAA,MAAM,IAAI,MAAM,kCAAkC,CAAA;AAAA,MACpD;AAEA,MAAA,MAAM,MAAA,GAAS,UAAU,SAAA,EAAU;AACnC,MAAA,MAAM,UAAA,GAAa,IAAIC,kBAAA,EAAY;AAGnC,MAAA,CAAC,YAAY;AACX,QAAA,IAAI;AACF,UAAA,OAAO,IAAA,EAAM;AACX,YAAA,MAAM,EAAE,IAAA,EAAM,KAAA,EAAM,GAAI,MAAM,OAAO,IAAA,EAAK;AAC1C,YAAA,IAAI,IAAA,EAAM;AACR,cAAA,UAAA,CAAW,GAAA,EAAI;AACf,cAAA;AAAA,YACF;AACA,YAAA,UAAA,CAAW,MAAM,KAAK,CAAA;AAAA,UACxB;AAAA,QACF,SAAS,KAAA,EAAO;AACd,UAAA,UAAA,CAAW,QAAQ,KAAc,CAAA;AAAA,QACnC;AAAA,MACF,CAAA,GAAG,CAAE,KAAA,CAAM,CAAA,KAAA,KAAS;AAClB,QAAA,UAAA,CAAW,QAAQ,KAAc,CAAA;AAAA,MACnC,CAAC,CAAA;AAED,MAAA,OAAO,UAAA;AAAA,IACT,CAAA,EAAG,sBAAsB,CAAA,EAAE;AAAA,EAC7B;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAOA,MAAM,WAAA,GAAc;AAClB,IAAA,OAAO,EAAE,SAAS,IAAA,EAAK;AAAA,EACzB;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAaA,MAAM,MAAA,CACJ,WAAA,EACA,OAAA,EAIc;AACd,IAAA,IAAI,CAAC,KAAK,eAAA,EAAiB;AACzB,MAAA,MAAM,IAAI,MAAM,0CAA0C,CAAA;AAAA,IAC5D;AAEA,IAAA,MAAM,SAAmB,EAAC;AAC1B,IAAA,WAAA,MAAiB,SAAS,WAAA,EAAa;AACrC,MAAA,IAAI,OAAO,UAAU,QAAA,EAAU;AAC7B,QAAA,MAAA,CAAO,IAAA,CAAK,MAAA,CAAO,IAAA,CAAK,KAAK,CAAC,CAAA;AAAA,MAChC,CAAA,MAAO;AACL,QAAA,MAAA,CAAO,KAAK,KAAK,CAAA;AAAA,MACnB;AAAA,IACF;AACA,IAAA,MAAM,MAAA,GAAS,MAAA,CAAO,MAAA,CAAO,MAAM,CAAA;AAEnC,IAAA,OAAO,IAAA,CAAK,OAAO,YAAY;AAC7B,MAAA,IAAI,CAAC,KAAK,eAAA,EAAiB;AACzB,QAAA,MAAM,IAAI,MAAM,gCAAgC,CAAA;AAAA,MAClD;AAEA,MAAA,MAAM,EAAE,SAAS,qBAAA,EAAuB,CAAA,EAAG,GAAG,WAAA,EAAY,GAAI,WAAW,EAAC;AAC1E,MAAA,MAAM,EAAE,MAAA,EAAQ,KAAA,EAAM,GAAI,MAAM,KAAK,eAAA,CAAgB,MAAA,CAAO,WAAA,CAAY,cAAA,CAAe,MAAA,EAAQ;AAAA,QAC7F,GAAG,WAAA;AAAA,QACH,KAAA,EAAO,KAAK,oBAAA,EAAsB,IAAA;AAAA,QAClC;AAAA,OACD,CAAA;AAED,MAAA,IAAI,KAAA,EAAO;AACT,QAAA,MAAM,KAAA;AAAA,MACR;AAEA,MAAA,MAAM,OAAA,GAAU,MAAA,CAAO,OAAA,EAAS,QAAA,GAAW,CAAC,CAAA;AAC5C,MAAA,MAAM,GAAA,GAKU,OAAA,EAAS,YAAA,GAAe,CAAC,CAAA;AAEzC,MAAA,IAAI,CAAC,GAAA,EAAK;AACR,QAAA,OAAO;AAAA,UACL,UAAA,EAAY,EAAA;AAAA,UACZ,OAAO,EAAC;AAAA,UACR,GAAA,EAAK;AAAA,SACP;AAAA,MACF;AAEA,MAAA,MAAM,QAAA,GAAgB;AAAA,QACpB,YAAY,GAAA,CAAI,UAAA;AAAA,QAChB,OAAO,GAAA,CAAI,KAAA;AAAA,QACX,GAAA,EAAK;AAAA,OACP;AAEA,MAAA,IAAI,OAAA,IAAW,IAAI,KAAA,EAAO;AACxB,QAAA,QAAA,CAAS,eAAA,GAAkB,GAAA,CAAI,KAAA,CAAM,GAAA,CAAI,CAAC,CAAA,MAAqB;AAAA,UAC7D,MAAM,CAAA,CAAE,IAAA;AAAA,UACR,SAAS,CAAA,CAAE,OAAA;AAAA,UACX,OAAO,CAAA,CAAE,KAAA;AAAA,UACT,KAAK,CAAA,CAAE;AAAA,SACT,CAAE,CAAA;AAAA,MACJ;AAEA,MAAA,OAAO,QAAA;AAAA,IACT,CAAA,EAAG,uBAAuB,CAAA,EAAE;AAAA,EAC9B;AACF","file":"index.cjs","sourcesContent":["/**\n * List of available Deepgram voice models for text-to-speech\n * Each voice is designed for specific use cases and languages\n * Format: {name}-{language} (e.g. asteria-en)\n */\nexport const DEEPGRAM_VOICES = [\n 'asteria-en',\n 'luna-en',\n 'stella-en',\n 'athena-en',\n 'hera-en',\n 'orion-en',\n 'arcas-en',\n 'perseus-en',\n 'angus-en',\n 'orpheus-en',\n 'helios-en',\n 'zeus-en',\n] as const;\n\nexport type DeepgramVoiceId = (typeof DEEPGRAM_VOICES)[number];\n\n/**\n * List of available Deepgram models for text-to-speech and speech-to-text\n */\nexport const DEEPGRAM_MODELS = ['aura', 'whisper', 'base', 'enhanced', 'nova', 'nova-2', 'nova-3'] as const;\n\nexport type DeepgramModel = (typeof DEEPGRAM_MODELS)[number];\n","import { PassThrough } from 'stream';\n\nimport { createClient } from '@deepgram/sdk';\nimport { MastraVoice } from '@mastra/core/voice';\n\nimport { DEEPGRAM_VOICES } from './voices';\nimport type { DeepgramVoiceId, DeepgramModel } from './voices';\n\ninterface DeepgramVoiceConfig {\n name?: DeepgramModel;\n apiKey?: string;\n properties?: Record<string, any>;\n language?: string;\n}\n\ninterface DeepgramWord {\n word: string;\n start?: number;\n end?: number;\n confidence?: number;\n speaker?: number;\n}\n\nexport class DeepgramVoice extends MastraVoice {\n private speechClient?: ReturnType<typeof createClient>;\n private listeningClient?: ReturnType<typeof createClient>;\n private storedSpeechModel?: { name: DeepgramModel; apiKey?: string };\n private storedListeningModel?: { name: DeepgramModel; apiKey?: string };\n private storedSpeaker?: DeepgramVoiceId;\n\n constructor({\n speechModel,\n listeningModel,\n speaker,\n }: { speechModel?: DeepgramVoiceConfig; listeningModel?: DeepgramVoiceConfig; speaker?: DeepgramVoiceId } = {}) {\n const defaultApiKey = process.env.DEEPGRAM_API_KEY;\n\n const defaultSpeechModel: { name: DeepgramModel; apiKey?: string } = {\n name: 'aura',\n apiKey: defaultApiKey,\n };\n\n const defaultListeningModel: { name: DeepgramModel; apiKey?: string } = {\n name: 'nova',\n apiKey: defaultApiKey,\n };\n\n super({\n speechModel: {\n name: speechModel?.name ?? defaultSpeechModel.name,\n apiKey: speechModel?.apiKey ?? defaultSpeechModel.apiKey,\n },\n listeningModel: {\n name: listeningModel?.name ?? defaultListeningModel.name,\n apiKey: listeningModel?.apiKey ?? defaultListeningModel.apiKey,\n },\n speaker,\n });\n\n this.storedSpeechModel = {\n name: speechModel?.name ?? defaultSpeechModel.name,\n apiKey: speechModel?.apiKey ?? defaultSpeechModel.apiKey,\n };\n this.storedListeningModel = {\n name: listeningModel?.name ?? defaultListeningModel.name,\n apiKey: listeningModel?.apiKey ?? defaultListeningModel.apiKey,\n };\n\n const speechApiKey = speechModel?.apiKey || defaultApiKey;\n const listeningApiKey = listeningModel?.apiKey || defaultApiKey;\n\n if (!speechApiKey && !listeningApiKey) {\n throw new Error('At least one of DEEPGRAM_API_KEY, speechModel.apiKey, or listeningModel.apiKey must be set');\n }\n\n if (speechApiKey) {\n this.speechClient = createClient(speechApiKey);\n }\n if (listeningApiKey) {\n this.listeningClient = createClient(listeningApiKey);\n }\n\n this.storedSpeaker = speaker || 'asteria-en';\n }\n\n async getSpeakers() {\n return this.traced(async () => {\n return DEEPGRAM_VOICES.map(voice => ({\n voiceId: voice,\n }));\n }, 'voice.deepgram.getSpeakers')();\n }\n\n async speak(\n input: string | NodeJS.ReadableStream,\n options?: {\n speaker?: string;\n [key: string]: any;\n },\n ): Promise<NodeJS.ReadableStream> {\n if (!this.speechClient) {\n throw new Error('Deepgram speech client not configured');\n }\n\n let text: string;\n if (typeof input !== 'string') {\n const chunks: Buffer[] = [];\n for await (const chunk of input) {\n if (typeof chunk === 'string') {\n chunks.push(Buffer.from(chunk));\n } else {\n chunks.push(chunk);\n }\n }\n text = Buffer.concat(chunks).toString('utf-8');\n } else {\n text = input;\n }\n\n if (text.trim().length === 0) {\n throw new Error('Input text is empty');\n }\n\n return this.traced(async () => {\n if (!this.speechClient) {\n throw new Error('No speech client configured');\n }\n\n const baseModel = this.storedSpeechModel?.name;\n const speakerId = options?.speaker || this.storedSpeaker;\n\n const modelName =\n baseModel && speakerId\n ? speakerId.startsWith(`${baseModel}-`)\n ? speakerId\n : `${baseModel}-${speakerId}`\n : baseModel || speakerId;\n\n const speakClient = this.speechClient.speak;\n const response = await speakClient.request(\n { text },\n {\n model: modelName,\n ...Object.fromEntries(Object.entries(options ?? {}).filter(([k]) => k !== 'speaker')),\n },\n );\n\n const webStream = await response.getStream();\n if (!webStream) {\n throw new Error('No stream returned from Deepgram');\n }\n\n const reader = webStream.getReader();\n const nodeStream = new PassThrough();\n\n // Add error handling for the stream processing\n (async () => {\n try {\n while (true) {\n const { done, value } = await reader.read();\n if (done) {\n nodeStream.end();\n break;\n }\n nodeStream.write(value);\n }\n } catch (error) {\n nodeStream.destroy(error as Error);\n }\n })().catch(error => {\n nodeStream.destroy(error as Error);\n });\n\n return nodeStream;\n }, 'voice.deepgram.speak')();\n }\n\n /**\n * Checks if listening capabilities are enabled.\n *\n * @returns {Promise<{ enabled: boolean }>}\n */\n async getListener() {\n return { enabled: true };\n }\n\n /**\n * Transcribes audio with optional speaker diarization.\n *\n * @param audioStream - Audio input stream\n * @param options - Transcription options (diarize, language, etc.)\n * @returns Promise resolving to:\n * - transcript: Full transcript string\n * - words: Array of word objects with timing and confidence\n * - raw: Complete Deepgram API response\n * - speakerSegments: (when diarize=true) Array of {word, speaker, start, end}\n */\n async listen(\n audioStream: NodeJS.ReadableStream,\n options?: {\n diarize?: boolean;\n [key: string]: any;\n },\n ): Promise<any> {\n if (!this.listeningClient) {\n throw new Error('Deepgram listening client not configured');\n }\n\n const chunks: Buffer[] = [];\n for await (const chunk of audioStream) {\n if (typeof chunk === 'string') {\n chunks.push(Buffer.from(chunk));\n } else {\n chunks.push(chunk);\n }\n }\n const buffer = Buffer.concat(chunks);\n\n return this.traced(async () => {\n if (!this.listeningClient) {\n throw new Error('No listening client configured');\n }\n\n const { diarize, diarize_speaker_count: _, ...restOptions } = options ?? {};\n const { result, error } = await this.listeningClient.listen.prerecorded.transcribeFile(buffer, {\n ...restOptions,\n model: this.storedListeningModel?.name,\n diarize,\n });\n\n if (error) {\n throw error;\n }\n\n const channel = result.results?.channels?.[0];\n const alt:\n | {\n transcript?: string;\n words?: DeepgramWord[];\n }\n | undefined = channel?.alternatives?.[0];\n\n if (!alt) {\n return {\n transcript: '',\n words: [],\n raw: result,\n };\n }\n\n const response: any = {\n transcript: alt.transcript,\n words: alt.words,\n raw: result,\n };\n\n if (diarize && alt.words) {\n response.speakerSegments = alt.words.map((w: DeepgramWord) => ({\n word: w.word,\n speaker: w.speaker,\n start: w.start,\n end: w.end,\n }));\n }\n\n return response;\n }, 'voice.deepgram.listen')();\n }\n}\n\nexport type { DeepgramVoiceConfig, DeepgramVoiceId, DeepgramModel };\n"]}
|
package/dist/index.d.ts
CHANGED
|
@@ -1,4 +1,52 @@
|
|
|
1
|
-
|
|
2
|
-
|
|
3
|
-
|
|
4
|
-
|
|
1
|
+
import { MastraVoice } from '@mastra/core/voice';
|
|
2
|
+
import type { DeepgramVoiceId, DeepgramModel } from './voices.js';
|
|
3
|
+
interface DeepgramVoiceConfig {
|
|
4
|
+
name?: DeepgramModel;
|
|
5
|
+
apiKey?: string;
|
|
6
|
+
properties?: Record<string, any>;
|
|
7
|
+
language?: string;
|
|
8
|
+
}
|
|
9
|
+
export declare class DeepgramVoice extends MastraVoice {
|
|
10
|
+
private speechClient?;
|
|
11
|
+
private listeningClient?;
|
|
12
|
+
private storedSpeechModel?;
|
|
13
|
+
private storedListeningModel?;
|
|
14
|
+
private storedSpeaker?;
|
|
15
|
+
constructor({ speechModel, listeningModel, speaker, }?: {
|
|
16
|
+
speechModel?: DeepgramVoiceConfig;
|
|
17
|
+
listeningModel?: DeepgramVoiceConfig;
|
|
18
|
+
speaker?: DeepgramVoiceId;
|
|
19
|
+
});
|
|
20
|
+
getSpeakers(): Promise<{
|
|
21
|
+
voiceId: "asteria-en" | "luna-en" | "stella-en" | "athena-en" | "hera-en" | "orion-en" | "arcas-en" | "perseus-en" | "angus-en" | "orpheus-en" | "helios-en" | "zeus-en";
|
|
22
|
+
}[]>;
|
|
23
|
+
speak(input: string | NodeJS.ReadableStream, options?: {
|
|
24
|
+
speaker?: string;
|
|
25
|
+
[key: string]: any;
|
|
26
|
+
}): Promise<NodeJS.ReadableStream>;
|
|
27
|
+
/**
|
|
28
|
+
* Checks if listening capabilities are enabled.
|
|
29
|
+
*
|
|
30
|
+
* @returns {Promise<{ enabled: boolean }>}
|
|
31
|
+
*/
|
|
32
|
+
getListener(): Promise<{
|
|
33
|
+
enabled: boolean;
|
|
34
|
+
}>;
|
|
35
|
+
/**
|
|
36
|
+
* Transcribes audio with optional speaker diarization.
|
|
37
|
+
*
|
|
38
|
+
* @param audioStream - Audio input stream
|
|
39
|
+
* @param options - Transcription options (diarize, language, etc.)
|
|
40
|
+
* @returns Promise resolving to:
|
|
41
|
+
* - transcript: Full transcript string
|
|
42
|
+
* - words: Array of word objects with timing and confidence
|
|
43
|
+
* - raw: Complete Deepgram API response
|
|
44
|
+
* - speakerSegments: (when diarize=true) Array of {word, speaker, start, end}
|
|
45
|
+
*/
|
|
46
|
+
listen(audioStream: NodeJS.ReadableStream, options?: {
|
|
47
|
+
diarize?: boolean;
|
|
48
|
+
[key: string]: any;
|
|
49
|
+
}): Promise<any>;
|
|
50
|
+
}
|
|
51
|
+
export type { DeepgramVoiceConfig, DeepgramVoiceId, DeepgramModel };
|
|
52
|
+
//# sourceMappingURL=index.d.ts.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAGA,OAAO,EAAE,WAAW,EAAE,MAAM,oBAAoB,CAAC;AAGjD,OAAO,KAAK,EAAE,eAAe,EAAE,aAAa,EAAE,MAAM,UAAU,CAAC;AAE/D,UAAU,mBAAmB;IAC3B,IAAI,CAAC,EAAE,aAAa,CAAC;IACrB,MAAM,CAAC,EAAE,MAAM,CAAC;IAChB,UAAU,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,GAAG,CAAC,CAAC;IACjC,QAAQ,CAAC,EAAE,MAAM,CAAC;CACnB;AAUD,qBAAa,aAAc,SAAQ,WAAW;IAC5C,OAAO,CAAC,YAAY,CAAC,CAAkC;IACvD,OAAO,CAAC,eAAe,CAAC,CAAkC;IAC1D,OAAO,CAAC,iBAAiB,CAAC,CAA2C;IACrE,OAAO,CAAC,oBAAoB,CAAC,CAA2C;IACxE,OAAO,CAAC,aAAa,CAAC,CAAkB;gBAE5B,EACV,WAAW,EACX,cAAc,EACd,OAAO,GACR,GAAE;QAAE,WAAW,CAAC,EAAE,mBAAmB,CAAC;QAAC,cAAc,CAAC,EAAE,mBAAmB,CAAC;QAAC,OAAO,CAAC,EAAE,eAAe,CAAA;KAAO;IAmDxG,WAAW;;;IAQX,KAAK,CACT,KAAK,EAAE,MAAM,GAAG,MAAM,CAAC,cAAc,EACrC,OAAO,CAAC,EAAE;QACR,OAAO,CAAC,EAAE,MAAM,CAAC;QACjB,CAAC,GAAG,EAAE,MAAM,GAAG,GAAG,CAAC;KACpB,GACA,OAAO,CAAC,MAAM,CAAC,cAAc,CAAC;IA8EjC;;;;OAIG;IACG,WAAW;;;IAIjB;;;;;;;;;;OAUG;IACG,MAAM,CACV,WAAW,EAAE,MAAM,CAAC,cAAc,EAClC,OAAO,CAAC,EAAE;QACR,OAAO,CAAC,EAAE,OAAO,CAAC;QAClB,CAAC,GAAG,EAAE,MAAM,GAAG,GAAG,CAAC;KACpB,GACA,OAAO,CAAC,GAAG,CAAC;CAiEhB;AAED,YAAY,EAAE,mBAAmB,EAAE,eAAe,EAAE,aAAa,EAAE,CAAC"}
|
package/dist/index.js
CHANGED
|
@@ -24,6 +24,9 @@ var DEEPGRAM_VOICES = [
|
|
|
24
24
|
var DeepgramVoice = class extends MastraVoice {
|
|
25
25
|
speechClient;
|
|
26
26
|
listeningClient;
|
|
27
|
+
storedSpeechModel;
|
|
28
|
+
storedListeningModel;
|
|
29
|
+
storedSpeaker;
|
|
27
30
|
constructor({
|
|
28
31
|
speechModel,
|
|
29
32
|
listeningModel,
|
|
@@ -49,6 +52,14 @@ var DeepgramVoice = class extends MastraVoice {
|
|
|
49
52
|
},
|
|
50
53
|
speaker
|
|
51
54
|
});
|
|
55
|
+
this.storedSpeechModel = {
|
|
56
|
+
name: speechModel?.name ?? defaultSpeechModel.name,
|
|
57
|
+
apiKey: speechModel?.apiKey ?? defaultSpeechModel.apiKey
|
|
58
|
+
};
|
|
59
|
+
this.storedListeningModel = {
|
|
60
|
+
name: listeningModel?.name ?? defaultListeningModel.name,
|
|
61
|
+
apiKey: listeningModel?.apiKey ?? defaultListeningModel.apiKey
|
|
62
|
+
};
|
|
52
63
|
const speechApiKey = speechModel?.apiKey || defaultApiKey;
|
|
53
64
|
const listeningApiKey = listeningModel?.apiKey || defaultApiKey;
|
|
54
65
|
if (!speechApiKey && !listeningApiKey) {
|
|
@@ -60,7 +71,7 @@ var DeepgramVoice = class extends MastraVoice {
|
|
|
60
71
|
if (listeningApiKey) {
|
|
61
72
|
this.listeningClient = createClient(listeningApiKey);
|
|
62
73
|
}
|
|
63
|
-
this.
|
|
74
|
+
this.storedSpeaker = speaker || "asteria-en";
|
|
64
75
|
}
|
|
65
76
|
async getSpeakers() {
|
|
66
77
|
return this.traced(async () => {
|
|
@@ -94,18 +105,15 @@ var DeepgramVoice = class extends MastraVoice {
|
|
|
94
105
|
if (!this.speechClient) {
|
|
95
106
|
throw new Error("No speech client configured");
|
|
96
107
|
}
|
|
97
|
-
|
|
98
|
-
|
|
99
|
-
|
|
100
|
-
} else if (this.speaker) {
|
|
101
|
-
model = this.speechModel?.name + "-" + this.speaker;
|
|
102
|
-
}
|
|
108
|
+
const baseModel = this.storedSpeechModel?.name;
|
|
109
|
+
const speakerId = options?.speaker || this.storedSpeaker;
|
|
110
|
+
const modelName = baseModel && speakerId ? speakerId.startsWith(`${baseModel}-`) ? speakerId : `${baseModel}-${speakerId}` : baseModel || speakerId;
|
|
103
111
|
const speakClient = this.speechClient.speak;
|
|
104
112
|
const response = await speakClient.request(
|
|
105
113
|
{ text },
|
|
106
114
|
{
|
|
107
|
-
model,
|
|
108
|
-
...options
|
|
115
|
+
model: modelName,
|
|
116
|
+
...Object.fromEntries(Object.entries(options ?? {}).filter(([k]) => k !== "speaker"))
|
|
109
117
|
}
|
|
110
118
|
);
|
|
111
119
|
const webStream = await response.getStream();
|
|
@@ -133,6 +141,25 @@ var DeepgramVoice = class extends MastraVoice {
|
|
|
133
141
|
return nodeStream;
|
|
134
142
|
}, "voice.deepgram.speak")();
|
|
135
143
|
}
|
|
144
|
+
/**
|
|
145
|
+
* Checks if listening capabilities are enabled.
|
|
146
|
+
*
|
|
147
|
+
* @returns {Promise<{ enabled: boolean }>}
|
|
148
|
+
*/
|
|
149
|
+
async getListener() {
|
|
150
|
+
return { enabled: true };
|
|
151
|
+
}
|
|
152
|
+
/**
|
|
153
|
+
* Transcribes audio with optional speaker diarization.
|
|
154
|
+
*
|
|
155
|
+
* @param audioStream - Audio input stream
|
|
156
|
+
* @param options - Transcription options (diarize, language, etc.)
|
|
157
|
+
* @returns Promise resolving to:
|
|
158
|
+
* - transcript: Full transcript string
|
|
159
|
+
* - words: Array of word objects with timing and confidence
|
|
160
|
+
* - raw: Complete Deepgram API response
|
|
161
|
+
* - speakerSegments: (when diarize=true) Array of {word, speaker, start, end}
|
|
162
|
+
*/
|
|
136
163
|
async listen(audioStream, options) {
|
|
137
164
|
if (!this.listeningClient) {
|
|
138
165
|
throw new Error("Deepgram listening client not configured");
|
|
@@ -150,20 +177,42 @@ var DeepgramVoice = class extends MastraVoice {
|
|
|
150
177
|
if (!this.listeningClient) {
|
|
151
178
|
throw new Error("No listening client configured");
|
|
152
179
|
}
|
|
180
|
+
const { diarize, diarize_speaker_count: _, ...restOptions } = options ?? {};
|
|
153
181
|
const { result, error } = await this.listeningClient.listen.prerecorded.transcribeFile(buffer, {
|
|
154
|
-
|
|
155
|
-
|
|
182
|
+
...restOptions,
|
|
183
|
+
model: this.storedListeningModel?.name,
|
|
184
|
+
diarize
|
|
156
185
|
});
|
|
157
186
|
if (error) {
|
|
158
187
|
throw error;
|
|
159
188
|
}
|
|
160
|
-
const
|
|
161
|
-
|
|
162
|
-
|
|
189
|
+
const channel = result.results?.channels?.[0];
|
|
190
|
+
const alt = channel?.alternatives?.[0];
|
|
191
|
+
if (!alt) {
|
|
192
|
+
return {
|
|
193
|
+
transcript: "",
|
|
194
|
+
words: [],
|
|
195
|
+
raw: result
|
|
196
|
+
};
|
|
197
|
+
}
|
|
198
|
+
const response = {
|
|
199
|
+
transcript: alt.transcript,
|
|
200
|
+
words: alt.words,
|
|
201
|
+
raw: result
|
|
202
|
+
};
|
|
203
|
+
if (diarize && alt.words) {
|
|
204
|
+
response.speakerSegments = alt.words.map((w) => ({
|
|
205
|
+
word: w.word,
|
|
206
|
+
speaker: w.speaker,
|
|
207
|
+
start: w.start,
|
|
208
|
+
end: w.end
|
|
209
|
+
}));
|
|
163
210
|
}
|
|
164
|
-
return
|
|
211
|
+
return response;
|
|
165
212
|
}, "voice.deepgram.listen")();
|
|
166
213
|
}
|
|
167
214
|
};
|
|
168
215
|
|
|
169
216
|
export { DeepgramVoice };
|
|
217
|
+
//# sourceMappingURL=index.js.map
|
|
218
|
+
//# sourceMappingURL=index.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"sources":["../src/voices.ts","../src/index.ts"],"names":[],"mappings":";;;;;;;AAKO,IAAM,eAAA,GAAkB;AAAA,EAC7B,YAAA;AAAA,EACA,SAAA;AAAA,EACA,WAAA;AAAA,EACA,WAAA;AAAA,EACA,SAAA;AAAA,EACA,UAAA;AAAA,EACA,UAAA;AAAA,EACA,YAAA;AAAA,EACA,UAAA;AAAA,EACA,YAAA;AAAA,EACA,WAAA;AAAA,EACA;AACF,CAAA;;;ACKO,IAAM,aAAA,GAAN,cAA4B,WAAA,CAAY;AAAA,EACrC,YAAA;AAAA,EACA,eAAA;AAAA,EACA,iBAAA;AAAA,EACA,oBAAA;AAAA,EACA,aAAA;AAAA,EAER,WAAA,CAAY;AAAA,IACV,WAAA;AAAA,IACA,cAAA;AAAA,IACA;AAAA,GACF,GAA4G,EAAC,EAAG;AAC9G,IAAA,MAAM,aAAA,GAAgB,QAAQ,GAAA,CAAI,gBAAA;AAElC,IAAA,MAAM,kBAAA,GAA+D;AAAA,MACnE,IAAA,EAAM,MAAA;AAAA,MACN,MAAA,EAAQ;AAAA,KACV;AAEA,IAAA,MAAM,qBAAA,GAAkE;AAAA,MACtE,IAAA,EAAM,MAAA;AAAA,MACN,MAAA,EAAQ;AAAA,KACV;AAEA,IAAA,KAAA,CAAM;AAAA,MACJ,WAAA,EAAa;AAAA,QACX,IAAA,EAAM,WAAA,EAAa,IAAA,IAAQ,kBAAA,CAAmB,IAAA;AAAA,QAC9C,MAAA,EAAQ,WAAA,EAAa,MAAA,IAAU,kBAAA,CAAmB;AAAA,OACpD;AAAA,MACA,cAAA,EAAgB;AAAA,QACd,IAAA,EAAM,cAAA,EAAgB,IAAA,IAAQ,qBAAA,CAAsB,IAAA;AAAA,QACpD,MAAA,EAAQ,cAAA,EAAgB,MAAA,IAAU,qBAAA,CAAsB;AAAA,OAC1D;AAAA,MACA;AAAA,KACD,CAAA;AAED,IAAA,IAAA,CAAK,iBAAA,GAAoB;AAAA,MACvB,IAAA,EAAM,WAAA,EAAa,IAAA,IAAQ,kBAAA,CAAmB,IAAA;AAAA,MAC9C,MAAA,EAAQ,WAAA,EAAa,MAAA,IAAU,kBAAA,CAAmB;AAAA,KACpD;AACA,IAAA,IAAA,CAAK,oBAAA,GAAuB;AAAA,MAC1B,IAAA,EAAM,cAAA,EAAgB,IAAA,IAAQ,qBAAA,CAAsB,IAAA;AAAA,MACpD,MAAA,EAAQ,cAAA,EAAgB,MAAA,IAAU,qBAAA,CAAsB;AAAA,KAC1D;AAEA,IAAA,MAAM,YAAA,GAAe,aAAa,MAAA,IAAU,aAAA;AAC5C,IAAA,MAAM,eAAA,GAAkB,gBAAgB,MAAA,IAAU,aAAA;AAElD,IAAA,IAAI,CAAC,YAAA,IAAgB,CAAC,eAAA,EAAiB;AACrC,MAAA,MAAM,IAAI,MAAM,4FAA4F,CAAA;AAAA,IAC9G;AAEA,IAAA,IAAI,YAAA,EAAc;AAChB,MAAA,IAAA,CAAK,YAAA,GAAe,aAAa,YAAY,CAAA;AAAA,IAC/C;AACA,IAAA,IAAI,eAAA,EAAiB;AACnB,MAAA,IAAA,CAAK,eAAA,GAAkB,aAAa,eAAe,CAAA;AAAA,IACrD;AAEA,IAAA,IAAA,CAAK,gBAAgB,OAAA,IAAW,YAAA;AAAA,EAClC;AAAA,EAEA,MAAM,WAAA,GAAc;AAClB,IAAA,OAAO,IAAA,CAAK,OAAO,YAAY;AAC7B,MAAA,OAAO,eAAA,CAAgB,IAAI,CAAA,KAAA,MAAU;AAAA,QACnC,OAAA,EAAS;AAAA,OACX,CAAE,CAAA;AAAA,IACJ,CAAA,EAAG,4BAA4B,CAAA,EAAE;AAAA,EACnC;AAAA,EAEA,MAAM,KAAA,CACJ,KAAA,EACA,OAAA,EAIgC;AAChC,IAAA,IAAI,CAAC,KAAK,YAAA,EAAc;AACtB,MAAA,MAAM,IAAI,MAAM,uCAAuC,CAAA;AAAA,IACzD;AAEA,IAAA,IAAI,IAAA;AACJ,IAAA,IAAI,OAAO,UAAU,QAAA,EAAU;AAC7B,MAAA,MAAM,SAAmB,EAAC;AAC1B,MAAA,WAAA,MAAiB,SAAS,KAAA,EAAO;AAC/B,QAAA,IAAI,OAAO,UAAU,QAAA,EAAU;AAC7B,UAAA,MAAA,CAAO,IAAA,CAAK,MAAA,CAAO,IAAA,CAAK,KAAK,CAAC,CAAA;AAAA,QAChC,CAAA,MAAO;AACL,UAAA,MAAA,CAAO,KAAK,KAAK,CAAA;AAAA,QACnB;AAAA,MACF;AACA,MAAA,IAAA,GAAO,MAAA,CAAO,MAAA,CAAO,MAAM,CAAA,CAAE,SAAS,OAAO,CAAA;AAAA,IAC/C,CAAA,MAAO;AACL,MAAA,IAAA,GAAO,KAAA;AAAA,IACT;AAEA,IAAA,IAAI,IAAA,CAAK,IAAA,EAAK,CAAE,MAAA,KAAW,CAAA,EAAG;AAC5B,MAAA,MAAM,IAAI,MAAM,qBAAqB,CAAA;AAAA,IACvC;AAEA,IAAA,OAAO,IAAA,CAAK,OAAO,YAAY;AAC7B,MAAA,IAAI,CAAC,KAAK,YAAA,EAAc;AACtB,QAAA,MAAM,IAAI,MAAM,6BAA6B,CAAA;AAAA,MAC/C;AAEA,MAAA,MAAM,SAAA,GAAY,KAAK,iBAAA,EAAmB,IAAA;AAC1C,MAAA,MAAM,SAAA,GAAY,OAAA,EAAS,OAAA,IAAW,IAAA,CAAK,aAAA;AAE3C,MAAA,MAAM,SAAA,GACJ,SAAA,IAAa,SAAA,GACT,SAAA,CAAU,WAAW,CAAA,EAAG,SAAS,CAAA,CAAA,CAAG,CAAA,GAClC,YACA,CAAA,EAAG,SAAS,CAAA,CAAA,EAAI,SAAS,KAC3B,SAAA,IAAa,SAAA;AAEnB,MAAA,MAAM,WAAA,GAAc,KAAK,YAAA,CAAa,KAAA;AACtC,MAAA,MAAM,QAAA,GAAW,MAAM,WAAA,CAAY,OAAA;AAAA,QACjC,EAAE,IAAA,EAAK;AAAA,QACP;AAAA,UACE,KAAA,EAAO,SAAA;AAAA,UACP,GAAG,MAAA,CAAO,WAAA,CAAY,MAAA,CAAO,OAAA,CAAQ,WAAW,EAAE,CAAA,CAAE,MAAA,CAAO,CAAC,CAAC,CAAC,CAAA,KAAM,CAAA,KAAM,SAAS,CAAC;AAAA;AACtF,OACF;AAEA,MAAA,MAAM,SAAA,GAAY,MAAM,QAAA,CAAS,SAAA,EAAU;AAC3C,MAAA,IAAI,CAAC,SAAA,EAAW;AACd,QAAA,MAAM,IAAI,MAAM,kCAAkC,CAAA;AAAA,MACpD;AAEA,MAAA,MAAM,MAAA,GAAS,UAAU,SAAA,EAAU;AACnC,MAAA,MAAM,UAAA,GAAa,IAAI,WAAA,EAAY;AAGnC,MAAA,CAAC,YAAY;AACX,QAAA,IAAI;AACF,UAAA,OAAO,IAAA,EAAM;AACX,YAAA,MAAM,EAAE,IAAA,EAAM,KAAA,EAAM,GAAI,MAAM,OAAO,IAAA,EAAK;AAC1C,YAAA,IAAI,IAAA,EAAM;AACR,cAAA,UAAA,CAAW,GAAA,EAAI;AACf,cAAA;AAAA,YACF;AACA,YAAA,UAAA,CAAW,MAAM,KAAK,CAAA;AAAA,UACxB;AAAA,QACF,SAAS,KAAA,EAAO;AACd,UAAA,UAAA,CAAW,QAAQ,KAAc,CAAA;AAAA,QACnC;AAAA,MACF,CAAA,GAAG,CAAE,KAAA,CAAM,CAAA,KAAA,KAAS;AAClB,QAAA,UAAA,CAAW,QAAQ,KAAc,CAAA;AAAA,MACnC,CAAC,CAAA;AAED,MAAA,OAAO,UAAA;AAAA,IACT,CAAA,EAAG,sBAAsB,CAAA,EAAE;AAAA,EAC7B;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAOA,MAAM,WAAA,GAAc;AAClB,IAAA,OAAO,EAAE,SAAS,IAAA,EAAK;AAAA,EACzB;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAaA,MAAM,MAAA,CACJ,WAAA,EACA,OAAA,EAIc;AACd,IAAA,IAAI,CAAC,KAAK,eAAA,EAAiB;AACzB,MAAA,MAAM,IAAI,MAAM,0CAA0C,CAAA;AAAA,IAC5D;AAEA,IAAA,MAAM,SAAmB,EAAC;AAC1B,IAAA,WAAA,MAAiB,SAAS,WAAA,EAAa;AACrC,MAAA,IAAI,OAAO,UAAU,QAAA,EAAU;AAC7B,QAAA,MAAA,CAAO,IAAA,CAAK,MAAA,CAAO,IAAA,CAAK,KAAK,CAAC,CAAA;AAAA,MAChC,CAAA,MAAO;AACL,QAAA,MAAA,CAAO,KAAK,KAAK,CAAA;AAAA,MACnB;AAAA,IACF;AACA,IAAA,MAAM,MAAA,GAAS,MAAA,CAAO,MAAA,CAAO,MAAM,CAAA;AAEnC,IAAA,OAAO,IAAA,CAAK,OAAO,YAAY;AAC7B,MAAA,IAAI,CAAC,KAAK,eAAA,EAAiB;AACzB,QAAA,MAAM,IAAI,MAAM,gCAAgC,CAAA;AAAA,MAClD;AAEA,MAAA,MAAM,EAAE,SAAS,qBAAA,EAAuB,CAAA,EAAG,GAAG,WAAA,EAAY,GAAI,WAAW,EAAC;AAC1E,MAAA,MAAM,EAAE,MAAA,EAAQ,KAAA,EAAM,GAAI,MAAM,KAAK,eAAA,CAAgB,MAAA,CAAO,WAAA,CAAY,cAAA,CAAe,MAAA,EAAQ;AAAA,QAC7F,GAAG,WAAA;AAAA,QACH,KAAA,EAAO,KAAK,oBAAA,EAAsB,IAAA;AAAA,QAClC;AAAA,OACD,CAAA;AAED,MAAA,IAAI,KAAA,EAAO;AACT,QAAA,MAAM,KAAA;AAAA,MACR;AAEA,MAAA,MAAM,OAAA,GAAU,MAAA,CAAO,OAAA,EAAS,QAAA,GAAW,CAAC,CAAA;AAC5C,MAAA,MAAM,GAAA,GAKU,OAAA,EAAS,YAAA,GAAe,CAAC,CAAA;AAEzC,MAAA,IAAI,CAAC,GAAA,EAAK;AACR,QAAA,OAAO;AAAA,UACL,UAAA,EAAY,EAAA;AAAA,UACZ,OAAO,EAAC;AAAA,UACR,GAAA,EAAK;AAAA,SACP;AAAA,MACF;AAEA,MAAA,MAAM,QAAA,GAAgB;AAAA,QACpB,YAAY,GAAA,CAAI,UAAA;AAAA,QAChB,OAAO,GAAA,CAAI,KAAA;AAAA,QACX,GAAA,EAAK;AAAA,OACP;AAEA,MAAA,IAAI,OAAA,IAAW,IAAI,KAAA,EAAO;AACxB,QAAA,QAAA,CAAS,eAAA,GAAkB,GAAA,CAAI,KAAA,CAAM,GAAA,CAAI,CAAC,CAAA,MAAqB;AAAA,UAC7D,MAAM,CAAA,CAAE,IAAA;AAAA,UACR,SAAS,CAAA,CAAE,OAAA;AAAA,UACX,OAAO,CAAA,CAAE,KAAA;AAAA,UACT,KAAK,CAAA,CAAE;AAAA,SACT,CAAE,CAAA;AAAA,MACJ;AAEA,MAAA,OAAO,QAAA;AAAA,IACT,CAAA,EAAG,uBAAuB,CAAA,EAAE;AAAA,EAC9B;AACF","file":"index.js","sourcesContent":["/**\n * List of available Deepgram voice models for text-to-speech\n * Each voice is designed for specific use cases and languages\n * Format: {name}-{language} (e.g. asteria-en)\n */\nexport const DEEPGRAM_VOICES = [\n 'asteria-en',\n 'luna-en',\n 'stella-en',\n 'athena-en',\n 'hera-en',\n 'orion-en',\n 'arcas-en',\n 'perseus-en',\n 'angus-en',\n 'orpheus-en',\n 'helios-en',\n 'zeus-en',\n] as const;\n\nexport type DeepgramVoiceId = (typeof DEEPGRAM_VOICES)[number];\n\n/**\n * List of available Deepgram models for text-to-speech and speech-to-text\n */\nexport const DEEPGRAM_MODELS = ['aura', 'whisper', 'base', 'enhanced', 'nova', 'nova-2', 'nova-3'] as const;\n\nexport type DeepgramModel = (typeof DEEPGRAM_MODELS)[number];\n","import { PassThrough } from 'stream';\n\nimport { createClient } from '@deepgram/sdk';\nimport { MastraVoice } from '@mastra/core/voice';\n\nimport { DEEPGRAM_VOICES } from './voices';\nimport type { DeepgramVoiceId, DeepgramModel } from './voices';\n\ninterface DeepgramVoiceConfig {\n name?: DeepgramModel;\n apiKey?: string;\n properties?: Record<string, any>;\n language?: string;\n}\n\ninterface DeepgramWord {\n word: string;\n start?: number;\n end?: number;\n confidence?: number;\n speaker?: number;\n}\n\nexport class DeepgramVoice extends MastraVoice {\n private speechClient?: ReturnType<typeof createClient>;\n private listeningClient?: ReturnType<typeof createClient>;\n private storedSpeechModel?: { name: DeepgramModel; apiKey?: string };\n private storedListeningModel?: { name: DeepgramModel; apiKey?: string };\n private storedSpeaker?: DeepgramVoiceId;\n\n constructor({\n speechModel,\n listeningModel,\n speaker,\n }: { speechModel?: DeepgramVoiceConfig; listeningModel?: DeepgramVoiceConfig; speaker?: DeepgramVoiceId } = {}) {\n const defaultApiKey = process.env.DEEPGRAM_API_KEY;\n\n const defaultSpeechModel: { name: DeepgramModel; apiKey?: string } = {\n name: 'aura',\n apiKey: defaultApiKey,\n };\n\n const defaultListeningModel: { name: DeepgramModel; apiKey?: string } = {\n name: 'nova',\n apiKey: defaultApiKey,\n };\n\n super({\n speechModel: {\n name: speechModel?.name ?? defaultSpeechModel.name,\n apiKey: speechModel?.apiKey ?? defaultSpeechModel.apiKey,\n },\n listeningModel: {\n name: listeningModel?.name ?? defaultListeningModel.name,\n apiKey: listeningModel?.apiKey ?? defaultListeningModel.apiKey,\n },\n speaker,\n });\n\n this.storedSpeechModel = {\n name: speechModel?.name ?? defaultSpeechModel.name,\n apiKey: speechModel?.apiKey ?? defaultSpeechModel.apiKey,\n };\n this.storedListeningModel = {\n name: listeningModel?.name ?? defaultListeningModel.name,\n apiKey: listeningModel?.apiKey ?? defaultListeningModel.apiKey,\n };\n\n const speechApiKey = speechModel?.apiKey || defaultApiKey;\n const listeningApiKey = listeningModel?.apiKey || defaultApiKey;\n\n if (!speechApiKey && !listeningApiKey) {\n throw new Error('At least one of DEEPGRAM_API_KEY, speechModel.apiKey, or listeningModel.apiKey must be set');\n }\n\n if (speechApiKey) {\n this.speechClient = createClient(speechApiKey);\n }\n if (listeningApiKey) {\n this.listeningClient = createClient(listeningApiKey);\n }\n\n this.storedSpeaker = speaker || 'asteria-en';\n }\n\n async getSpeakers() {\n return this.traced(async () => {\n return DEEPGRAM_VOICES.map(voice => ({\n voiceId: voice,\n }));\n }, 'voice.deepgram.getSpeakers')();\n }\n\n async speak(\n input: string | NodeJS.ReadableStream,\n options?: {\n speaker?: string;\n [key: string]: any;\n },\n ): Promise<NodeJS.ReadableStream> {\n if (!this.speechClient) {\n throw new Error('Deepgram speech client not configured');\n }\n\n let text: string;\n if (typeof input !== 'string') {\n const chunks: Buffer[] = [];\n for await (const chunk of input) {\n if (typeof chunk === 'string') {\n chunks.push(Buffer.from(chunk));\n } else {\n chunks.push(chunk);\n }\n }\n text = Buffer.concat(chunks).toString('utf-8');\n } else {\n text = input;\n }\n\n if (text.trim().length === 0) {\n throw new Error('Input text is empty');\n }\n\n return this.traced(async () => {\n if (!this.speechClient) {\n throw new Error('No speech client configured');\n }\n\n const baseModel = this.storedSpeechModel?.name;\n const speakerId = options?.speaker || this.storedSpeaker;\n\n const modelName =\n baseModel && speakerId\n ? speakerId.startsWith(`${baseModel}-`)\n ? speakerId\n : `${baseModel}-${speakerId}`\n : baseModel || speakerId;\n\n const speakClient = this.speechClient.speak;\n const response = await speakClient.request(\n { text },\n {\n model: modelName,\n ...Object.fromEntries(Object.entries(options ?? {}).filter(([k]) => k !== 'speaker')),\n },\n );\n\n const webStream = await response.getStream();\n if (!webStream) {\n throw new Error('No stream returned from Deepgram');\n }\n\n const reader = webStream.getReader();\n const nodeStream = new PassThrough();\n\n // Add error handling for the stream processing\n (async () => {\n try {\n while (true) {\n const { done, value } = await reader.read();\n if (done) {\n nodeStream.end();\n break;\n }\n nodeStream.write(value);\n }\n } catch (error) {\n nodeStream.destroy(error as Error);\n }\n })().catch(error => {\n nodeStream.destroy(error as Error);\n });\n\n return nodeStream;\n }, 'voice.deepgram.speak')();\n }\n\n /**\n * Checks if listening capabilities are enabled.\n *\n * @returns {Promise<{ enabled: boolean }>}\n */\n async getListener() {\n return { enabled: true };\n }\n\n /**\n * Transcribes audio with optional speaker diarization.\n *\n * @param audioStream - Audio input stream\n * @param options - Transcription options (diarize, language, etc.)\n * @returns Promise resolving to:\n * - transcript: Full transcript string\n * - words: Array of word objects with timing and confidence\n * - raw: Complete Deepgram API response\n * - speakerSegments: (when diarize=true) Array of {word, speaker, start, end}\n */\n async listen(\n audioStream: NodeJS.ReadableStream,\n options?: {\n diarize?: boolean;\n [key: string]: any;\n },\n ): Promise<any> {\n if (!this.listeningClient) {\n throw new Error('Deepgram listening client not configured');\n }\n\n const chunks: Buffer[] = [];\n for await (const chunk of audioStream) {\n if (typeof chunk === 'string') {\n chunks.push(Buffer.from(chunk));\n } else {\n chunks.push(chunk);\n }\n }\n const buffer = Buffer.concat(chunks);\n\n return this.traced(async () => {\n if (!this.listeningClient) {\n throw new Error('No listening client configured');\n }\n\n const { diarize, diarize_speaker_count: _, ...restOptions } = options ?? {};\n const { result, error } = await this.listeningClient.listen.prerecorded.transcribeFile(buffer, {\n ...restOptions,\n model: this.storedListeningModel?.name,\n diarize,\n });\n\n if (error) {\n throw error;\n }\n\n const channel = result.results?.channels?.[0];\n const alt:\n | {\n transcript?: string;\n words?: DeepgramWord[];\n }\n | undefined = channel?.alternatives?.[0];\n\n if (!alt) {\n return {\n transcript: '',\n words: [],\n raw: result,\n };\n }\n\n const response: any = {\n transcript: alt.transcript,\n words: alt.words,\n raw: result,\n };\n\n if (diarize && alt.words) {\n response.speakerSegments = alt.words.map((w: DeepgramWord) => ({\n word: w.word,\n speaker: w.speaker,\n start: w.start,\n end: w.end,\n }));\n }\n\n return response;\n }, 'voice.deepgram.listen')();\n }\n}\n\nexport type { DeepgramVoiceConfig, DeepgramVoiceId, DeepgramModel };\n"]}
|
package/dist/voices.d.ts
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* List of available Deepgram voice models for text-to-speech
|
|
3
|
+
* Each voice is designed for specific use cases and languages
|
|
4
|
+
* Format: {name}-{language} (e.g. asteria-en)
|
|
5
|
+
*/
|
|
6
|
+
export declare const DEEPGRAM_VOICES: readonly ["asteria-en", "luna-en", "stella-en", "athena-en", "hera-en", "orion-en", "arcas-en", "perseus-en", "angus-en", "orpheus-en", "helios-en", "zeus-en"];
|
|
7
|
+
export type DeepgramVoiceId = (typeof DEEPGRAM_VOICES)[number];
|
|
8
|
+
/**
|
|
9
|
+
* List of available Deepgram models for text-to-speech and speech-to-text
|
|
10
|
+
*/
|
|
11
|
+
export declare const DEEPGRAM_MODELS: readonly ["aura", "whisper", "base", "enhanced", "nova", "nova-2", "nova-3"];
|
|
12
|
+
export type DeepgramModel = (typeof DEEPGRAM_MODELS)[number];
|
|
13
|
+
//# sourceMappingURL=voices.d.ts.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"voices.d.ts","sourceRoot":"","sources":["../src/voices.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AACH,eAAO,MAAM,eAAe,iKAalB,CAAC;AAEX,MAAM,MAAM,eAAe,GAAG,CAAC,OAAO,eAAe,CAAC,CAAC,MAAM,CAAC,CAAC;AAE/D;;GAEG;AACH,eAAO,MAAM,eAAe,8EAA+E,CAAC;AAE5G,MAAM,MAAM,aAAa,GAAG,CAAC,OAAO,eAAe,CAAC,CAAC,MAAM,CAAC,CAAC"}
|
package/package.json
CHANGED
|
@@ -1,10 +1,11 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@mastra/voice-deepgram",
|
|
3
|
-
"version": "0.0.0-
|
|
3
|
+
"version": "0.0.0-safe-stringify-telemetry-20251205024938",
|
|
4
4
|
"description": "Mastra Deepgram voice integration",
|
|
5
5
|
"type": "module",
|
|
6
6
|
"files": [
|
|
7
|
-
"dist"
|
|
7
|
+
"dist",
|
|
8
|
+
"CHANGELOG.md"
|
|
8
9
|
],
|
|
9
10
|
"main": "dist/index.js",
|
|
10
11
|
"types": "dist/index.d.ts",
|
|
@@ -15,30 +16,50 @@
|
|
|
15
16
|
"default": "./dist/index.js"
|
|
16
17
|
},
|
|
17
18
|
"require": {
|
|
18
|
-
"types": "./dist/index.d.
|
|
19
|
+
"types": "./dist/index.d.ts",
|
|
19
20
|
"default": "./dist/index.cjs"
|
|
20
21
|
}
|
|
21
22
|
},
|
|
22
23
|
"./package.json": "./package.json"
|
|
23
24
|
},
|
|
24
|
-
"license": "
|
|
25
|
+
"license": "Apache-2.0",
|
|
25
26
|
"dependencies": {
|
|
26
|
-
"@deepgram/sdk": "^3.
|
|
27
|
-
"zod": "^3.24.2",
|
|
28
|
-
"@mastra/core": "0.0.0-revert-schema-20250416221206"
|
|
27
|
+
"@deepgram/sdk": "^3.13.0"
|
|
29
28
|
},
|
|
30
29
|
"devDependencies": {
|
|
31
|
-
"@microsoft/api-extractor": "^7.52.
|
|
32
|
-
"@types/node": "^20.
|
|
33
|
-
"eslint": "^9.
|
|
34
|
-
"tsup": "^8.
|
|
35
|
-
"typescript": "^5.8.
|
|
36
|
-
"vitest": "^2.
|
|
37
|
-
"@internal/lint": "0.0.
|
|
30
|
+
"@microsoft/api-extractor": "^7.52.8",
|
|
31
|
+
"@types/node": "^20.19.0",
|
|
32
|
+
"eslint": "^9.37.0",
|
|
33
|
+
"tsup": "^8.5.0",
|
|
34
|
+
"typescript": "^5.8.3",
|
|
35
|
+
"vitest": "^3.2.4",
|
|
36
|
+
"@internal/lint": "0.0.0-safe-stringify-telemetry-20251205024938",
|
|
37
|
+
"@internal/types-builder": "0.0.0-safe-stringify-telemetry-20251205024938",
|
|
38
|
+
"@mastra/core": "0.0.0-safe-stringify-telemetry-20251205024938"
|
|
39
|
+
},
|
|
40
|
+
"peerDependencies": {
|
|
41
|
+
"zod": "^3.25.0 || ^4.0.0",
|
|
42
|
+
"@mastra/core": "0.0.0-safe-stringify-telemetry-20251205024938"
|
|
43
|
+
},
|
|
44
|
+
"homepage": "https://mastra.ai",
|
|
45
|
+
"repository": {
|
|
46
|
+
"type": "git",
|
|
47
|
+
"url": "git+https://github.com/mastra-ai/mastra.git",
|
|
48
|
+
"directory": "voice/deepgram"
|
|
49
|
+
},
|
|
50
|
+
"bugs": {
|
|
51
|
+
"url": "https://github.com/mastra-ai/mastra/issues"
|
|
52
|
+
},
|
|
53
|
+
"publishConfig": {
|
|
54
|
+
"access": "public",
|
|
55
|
+
"publish-branch": [
|
|
56
|
+
"main",
|
|
57
|
+
"0.x"
|
|
58
|
+
]
|
|
38
59
|
},
|
|
39
60
|
"scripts": {
|
|
40
|
-
"build": "tsup
|
|
41
|
-
"build:watch": "
|
|
61
|
+
"build": "tsup --silent --config tsup.config.ts",
|
|
62
|
+
"build:watch": "tsup --watch --silent --config tsup.config.ts",
|
|
42
63
|
"test": "vitest run",
|
|
43
64
|
"lint": "eslint ."
|
|
44
65
|
}
|
|
@@ -1,50 +0,0 @@
|
|
|
1
|
-
import { MastraVoice } from '@mastra/core/voice';
|
|
2
|
-
|
|
3
|
-
/**
|
|
4
|
-
* List of available Deepgram models for text-to-speech and speech-to-text
|
|
5
|
-
*/
|
|
6
|
-
export declare const DEEPGRAM_MODELS: readonly ["aura", "whisper", "base", "enhanced", "nova", "nova-2", "nova-3"];
|
|
7
|
-
|
|
8
|
-
/**
|
|
9
|
-
* List of available Deepgram voice models for text-to-speech
|
|
10
|
-
* Each voice is designed for specific use cases and languages
|
|
11
|
-
* Format: {name}-{language} (e.g. asteria-en)
|
|
12
|
-
*/
|
|
13
|
-
export declare const DEEPGRAM_VOICES: readonly ["asteria-en", "luna-en", "stella-en", "athena-en", "hera-en", "orion-en", "arcas-en", "perseus-en", "angus-en", "orpheus-en", "helios-en", "zeus-en"];
|
|
14
|
-
|
|
15
|
-
declare type DeepgramModel = (typeof DEEPGRAM_MODELS)[number];
|
|
16
|
-
export { DeepgramModel }
|
|
17
|
-
export { DeepgramModel as DeepgramModel_alias_1 }
|
|
18
|
-
|
|
19
|
-
export declare class DeepgramVoice extends MastraVoice {
|
|
20
|
-
private speechClient?;
|
|
21
|
-
private listeningClient?;
|
|
22
|
-
constructor({ speechModel, listeningModel, speaker, }?: {
|
|
23
|
-
speechModel?: DeepgramVoiceConfig;
|
|
24
|
-
listeningModel?: DeepgramVoiceConfig;
|
|
25
|
-
speaker?: DeepgramVoiceId;
|
|
26
|
-
});
|
|
27
|
-
getSpeakers(): Promise<{
|
|
28
|
-
voiceId: "asteria-en" | "luna-en" | "stella-en" | "athena-en" | "hera-en" | "orion-en" | "arcas-en" | "perseus-en" | "angus-en" | "orpheus-en" | "helios-en" | "zeus-en";
|
|
29
|
-
}[]>;
|
|
30
|
-
speak(input: string | NodeJS.ReadableStream, options?: {
|
|
31
|
-
speaker?: string;
|
|
32
|
-
[key: string]: any;
|
|
33
|
-
}): Promise<NodeJS.ReadableStream>;
|
|
34
|
-
listen(audioStream: NodeJS.ReadableStream, options?: {
|
|
35
|
-
[key: string]: any;
|
|
36
|
-
}): Promise<string>;
|
|
37
|
-
}
|
|
38
|
-
|
|
39
|
-
export declare interface DeepgramVoiceConfig {
|
|
40
|
-
name?: DeepgramModel;
|
|
41
|
-
apiKey?: string;
|
|
42
|
-
properties?: Record<string, any>;
|
|
43
|
-
language?: string;
|
|
44
|
-
}
|
|
45
|
-
|
|
46
|
-
declare type DeepgramVoiceId = (typeof DEEPGRAM_VOICES)[number];
|
|
47
|
-
export { DeepgramVoiceId }
|
|
48
|
-
export { DeepgramVoiceId as DeepgramVoiceId_alias_1 }
|
|
49
|
-
|
|
50
|
-
export { }
|