@vocalstack/mcp 1.0.0 → 1.3.0

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.
@@ -0,0 +1,708 @@
1
+ {
2
+ "instructions": "VocalStack transcribes and translates recorded or live audio and video in about 100 languages, with speaker labels.\nRecorded audio: transcribe (a link or a file) returns a transcriptionId at once; get_transcription with waitS waits for it and reads the text, in any language with lang.\nFind: list_transcriptions (ids, newest first); search_transcriptions (mode words for words or names, meaning for a question).\nOne transcription: get_summary; rename_speaker; set_cloned_voices; delete_transcription and restore_transcription.\nLive streams: start_broadcast, then get_transcription while it runs, stop_broadcast to end it; list_broadcasts shows them.\nPeople: list_speakers, save_speaker (voice cloned only with the person's consent), delete_speaker.\nOnly transcription is billed; everything else, and listening, is free. Retry a start with the same idempotencyKey.\nErrors are results with isError true: error, message, retryable, and retryAfterS or topUpUrl when they apply.",
3
+ "tools": [
4
+ {
5
+ "name": "transcribe",
6
+ "title": "Transcribe a recording",
7
+ "description": "Start transcribing a finished recording, from a public link or a file sent inline; returns a transcriptionId at once, before any text.\nUse: audio or video that has ended (a file, podcast episode, YouTube video or public post). A stream playing now: start_broadcast. The result: get_transcription with waitS, no polling needed.\nList price: $0.20 an hour of audio, billed per second transcribed; Pay as you go bills at least 5 minutes per transcription. Each call is a new, billed transcription unless it repeats an idempotencyKey; a refused call costs nothing; summary and shareLink are free.\nNew transcriptions an hour per account: 10 on Pay as you go, 100 on Premium.\nTranscriptions running at once per account: 5 on Pay as you go, 50 on Premium; more wait their turn instead of failing.\nReturns: transcriptionId, state (queued, fetching, transcribing, then done, failed or stopped), progress, shareUrl when shareLink is true, replayed true for a repeated key.\nErrors: invalid_options (neither or both of url and fileBase64, not a public http(s) link, bad base64, too large), invalid_audio (empty file), idempotency_key_reused, insufficient_credit (with topUpUrl), quota_exceeded, rate_limited (with retryAfterS). A source that cannot be read ends in state failed, with error.",
8
+ "inputSchema": {
9
+ "additionalProperties": false,
10
+ "properties": {
11
+ "endS": {
12
+ "description": "Transcribe only up to this second (default: the end), at least 0.5 s after startS.",
13
+ "examples": [
14
+ 600
15
+ ],
16
+ "type": "number"
17
+ },
18
+ "fileBase64": {
19
+ "contentEncoding": "base64",
20
+ "description": "The whole file in standard base64 (a data: URL prefix and line breaks are accepted), at most 10 MiB decoded; for a larger file give url instead. mp3, m4a, wav, mp4, webm and other common formats; the format is read from the file. Needs filename.",
21
+ "type": "string"
22
+ },
23
+ "filename": {
24
+ "description": "With fileBase64 only: the file's name with its extension, also the default display name.",
25
+ "examples": [
26
+ "meeting.m4a"
27
+ ],
28
+ "maxLength": 255,
29
+ "type": "string"
30
+ },
31
+ "idempotencyKey": {
32
+ "description": "Any unique string of visible ASCII. Retrying with the same key and the same arguments within 24 h returns the first result (with replayed true) and nothing is created or billed twice; the same key with other arguments is refused.",
33
+ "examples": [
34
+ "meeting-2026-11-19"
35
+ ],
36
+ "maxLength": 255,
37
+ "type": "string"
38
+ },
39
+ "mimeType": {
40
+ "description": "With fileBase64 only, optional: the file's media type when known.",
41
+ "examples": [
42
+ "audio/mpeg",
43
+ "video/mp4"
44
+ ],
45
+ "type": "string"
46
+ },
47
+ "name": {
48
+ "description": "Display name shown in lists and on the share page. Defaults to the file name or the link.",
49
+ "examples": [
50
+ "Weekly team call"
51
+ ],
52
+ "maxLength": 200,
53
+ "type": "string"
54
+ },
55
+ "shareLink": {
56
+ "default": false,
57
+ "description": "Make a public link anyone can open to read the transcript without signing in (shareUrl in the result). Off: it stays private to this account.",
58
+ "type": "boolean"
59
+ },
60
+ "speakerLabels": {
61
+ "default": "auto",
62
+ "description": "auto labels lines once a second speaker is heard, on labels every line, off none. Recordings under a minute are never labelled.",
63
+ "enum": [
64
+ "auto",
65
+ "on",
66
+ "off"
67
+ ]
68
+ },
69
+ "spokenLanguage": {
70
+ "description": "Code of the language spoken, e.g. en, ro, es. Omit to detect it (recordings that switch language are handled too); set it for better accuracy on short or noisy audio.",
71
+ "examples": [
72
+ "en",
73
+ "es"
74
+ ],
75
+ "type": "string"
76
+ },
77
+ "startS": {
78
+ "description": "Transcribe only from this second (default: the start). Only the part transcribed is billed: startS 90 and endS 600 bill 8.5 minutes.",
79
+ "examples": [
80
+ 90
81
+ ],
82
+ "minimum": 0,
83
+ "type": "number"
84
+ },
85
+ "summary": {
86
+ "default": false,
87
+ "description": "Also write a title and summary when it is done, at no extra cost (get_summary reads it). Never for recordings under a minute.",
88
+ "type": "boolean"
89
+ },
90
+ "url": {
91
+ "description": "Public http(s) link to the recording: an audio or video file, a YouTube video, or a public post on Twitch, Facebook, Instagram, Vimeo, Dailymotion, TikTok, X, SoundCloud, Kick, Mixcloud, Bandcamp, LinkedIn, TED, Internet Archive, Streamable, Odysee, Niconico, VK, BitChute, Wistia, Vidyard, Kaltura, JW Player, Pinterest, Bluesky, Google Drive, Dropbox, Spreaker, iHeartRadio, Audioboom, CNN or Loom. It must open without signing in and must not be a live stream. Give either url or fileBase64, not both.",
92
+ "examples": [
93
+ "https://www.youtube.com/watch?v=jNQXAC9IVRw",
94
+ "https://example.com/episode-12.mp3"
95
+ ],
96
+ "format": "uri",
97
+ "maxLength": 2048,
98
+ "type": "string"
99
+ }
100
+ },
101
+ "type": "object"
102
+ },
103
+ "annotations": {
104
+ "title": "Transcribe a recording",
105
+ "readOnlyHint": false,
106
+ "destructiveHint": false,
107
+ "idempotentHint": false,
108
+ "openWorldHint": true
109
+ }
110
+ },
111
+ {
112
+ "name": "get_transcription",
113
+ "title": "Read a transcription",
114
+ "description": "Read one transcription's status and text, with timestamps and speakers: as it was spoken or translated, at once or after waiting for it to finish.\nUse: after transcribe (waitS saves polling), or any time during a live transcription for the text so far. An overview: get_summary. Finding a passage: search_transcriptions.\nBehaviour: free. A wait ends as soon as the work does, and sends progress notifications when the client gives a progress token. lang needs a finished transcription; the first call in a new language can take up to a minute on a long one, later pages are quick. Page with offset = nextOffset.\nReturns: status (state, progress, error when failed), timedOut when waitS was set, then segmentCount, partial while text is still coming, text or segments, nextOffset when more remain, and lang and translated with lang. A wait that runs out with lang set returns the status only.\nErrors: not_found; invalid_state for lang before it is finished; invalid_options for an unsupported lang or includeWords with lang; rate_limited when waitS is set and 4 waits are already open for this key (8 for the account).",
115
+ "inputSchema": {
116
+ "additionalProperties": false,
117
+ "properties": {
118
+ "format": {
119
+ "default": "text",
120
+ "description": "text: one string, a line per segment as \"[mm:ss] Speaker 1: words\", for reading and quoting; segments: timed lines (start, end, text, speakerNumber, speakerName), for subtitles and exact times; none: status only (with waitS: wait without reading).",
121
+ "enum": [
122
+ "text",
123
+ "segments",
124
+ "none"
125
+ ]
126
+ },
127
+ "includeWords": {
128
+ "default": false,
129
+ "description": "With format segments and no lang: also each word with its start and end second. It makes the answer several times larger, so ask for a narrow page (e.g. limit 50).",
130
+ "type": "boolean"
131
+ },
132
+ "lang": {
133
+ "description": "Code of a language to read the whole text in, one of about 100 (e.g. es, de, ja): the transcription must be finished (state done, or a stopped live one). The language already spoken returns the original with translated false. Omit for the original.",
134
+ "examples": [
135
+ "es",
136
+ "ja"
137
+ ],
138
+ "type": "string"
139
+ },
140
+ "limit": {
141
+ "default": 500,
142
+ "description": "Most segments in one call. When more are left the result has nextOffset.",
143
+ "examples": [
144
+ 200
145
+ ],
146
+ "maximum": 2000,
147
+ "minimum": 1,
148
+ "type": "integer"
149
+ },
150
+ "offset": {
151
+ "default": 0,
152
+ "description": "Index of the first segment to return (0, the start, by default). To page, pass the nextOffset the previous call returned.",
153
+ "examples": [
154
+ 500
155
+ ],
156
+ "minimum": 0,
157
+ "type": "integer"
158
+ },
159
+ "transcriptionId": {
160
+ "description": "A transcriptionId from transcribe, list_transcriptions, search_transcriptions or start_broadcast.",
161
+ "examples": [
162
+ "0190f0e0-7c1a-7d3e-9b2a-5f4e3d2c1b0a"
163
+ ],
164
+ "format": "uuid",
165
+ "type": "string"
166
+ },
167
+ "waitS": {
168
+ "default": 0,
169
+ "description": "Seconds to wait for a file or link transcription to finish before answering, 0 to 60 (0, the default: answer at once). Returns as soon as it ends; timedOut true means it is still running, so call again. Keep it under your client's call timeout. A live transcription never waits.",
170
+ "examples": [
171
+ 50
172
+ ],
173
+ "maximum": 60,
174
+ "minimum": 0,
175
+ "type": "integer"
176
+ }
177
+ },
178
+ "required": [
179
+ "transcriptionId"
180
+ ],
181
+ "type": "object"
182
+ },
183
+ "annotations": {
184
+ "title": "Read a transcription",
185
+ "readOnlyHint": true,
186
+ "destructiveHint": false,
187
+ "idempotentHint": true,
188
+ "openWorldHint": false
189
+ }
190
+ },
191
+ {
192
+ "name": "list_transcriptions",
193
+ "title": "List transcriptions",
194
+ "description": "List this account's transcriptions, newest first, a page at a time: files, links and live ones, including those broadcasts recorded.\nUse: to find a transcriptionId or see what exists. What was said: search_transcriptions. Broadcasts: list_broadcasts.\nBehaviour: free. lifecycle archived lists the trash, which restore_transcription can empty back for 30 days. A cursor this tool did not give out is refused.\nReturns: transcriptions (transcriptionId, name, createdAt, kind file or live, state, durationS and title when known, broadcastId for a broadcast's), and nextCursor while more remain. None: an empty list.\nErrors: invalid_options for an unknown cursor.",
195
+ "inputSchema": {
196
+ "additionalProperties": false,
197
+ "properties": {
198
+ "cursor": {
199
+ "description": "Omit for the first page (the newest). For the next page, pass the previous call's nextCursor unchanged, with the same lifecycle.",
200
+ "type": "string"
201
+ },
202
+ "lifecycle": {
203
+ "default": "active",
204
+ "description": "Which transcriptions: active those in use, archived those in the trash, all both.",
205
+ "enum": [
206
+ "active",
207
+ "archived",
208
+ "all"
209
+ ]
210
+ },
211
+ "limit": {
212
+ "default": 20,
213
+ "description": "Most transcriptions in one page.",
214
+ "examples": [
215
+ 50
216
+ ],
217
+ "maximum": 100,
218
+ "minimum": 1,
219
+ "type": "integer"
220
+ }
221
+ },
222
+ "type": "object"
223
+ },
224
+ "annotations": {
225
+ "title": "List transcriptions",
226
+ "readOnlyHint": true,
227
+ "destructiveHint": false,
228
+ "idempotentHint": true,
229
+ "openWorldHint": false
230
+ }
231
+ },
232
+ {
233
+ "name": "search_transcriptions",
234
+ "title": "Search transcriptions",
235
+ "description": "Find what was said across this account's transcriptions, or in one: the lines with given words or a name (mode words), or the passages that answer a question (mode meaning).\nUse: words when you know what was said; meaning for a question or topic, in any language. Text around a hit: get_transcription with offset. One transcription's overview: get_summary.\nBehaviour: free; no paging. words ranks whole words first, then close spellings, then newest, and searches the query's English translation too. meaning returns up to 10 passages, closest first. Searches by meaning an hour per account: 60 on Pay as you go, no limit on Premium.\nReturns: words: query, queryEn, hits (transcriptionId, name, mediaStart in seconds, speaker, text, score, live, recordedAt, rendered with lang). meaning: query, state (ready, or unavailable or off when it cannot run), passages (transcriptionId, name, mediaStart, mediaEnd, text, similarity 0 to 1, live), searchesLeftThisHour when capped. Cite the transcription and second. No match: an empty list.\nErrors: not_found for an unknown transcriptionId; invalid_options for lang or limit with mode meaning; rate_limited (with retryAfterS) past the hourly searches by meaning.",
236
+ "inputSchema": {
237
+ "additionalProperties": false,
238
+ "properties": {
239
+ "lang": {
240
+ "description": "mode words only, optional language code: each hit also comes back translated into it, as rendered. It never changes what matches.",
241
+ "examples": [
242
+ "en",
243
+ "de"
244
+ ],
245
+ "type": "string"
246
+ },
247
+ "limit": {
248
+ "default": 20,
249
+ "description": "mode words only: most lines to return, best first. There is no next page: narrow the query or set transcriptionId.",
250
+ "examples": [
251
+ 10
252
+ ],
253
+ "maximum": 50,
254
+ "minimum": 1,
255
+ "type": "integer"
256
+ },
257
+ "mode": {
258
+ "default": "words",
259
+ "description": "words: the lines where these words or a close spelling were said (\"presupuesto\" also finds \"budget\"). meaning: the passages closest in meaning, even with none of the words.",
260
+ "enum": [
261
+ "words",
262
+ "meaning"
263
+ ]
264
+ },
265
+ "query": {
266
+ "description": "mode words: words, a phrase or a name that was said. mode meaning: a question or topic in plain words; a full question works better than one keyword. Any language.",
267
+ "examples": [
268
+ "quarterly budget",
269
+ "What did they decide about hiring?"
270
+ ],
271
+ "maxLength": 200,
272
+ "minLength": 2,
273
+ "type": "string"
274
+ },
275
+ "transcriptionId": {
276
+ "description": "Search only this transcription (a transcriptionId from list_transcriptions or transcribe). Omit to search the whole account.",
277
+ "examples": [
278
+ "0190f0e0-7c1a-7d3e-9b2a-5f4e3d2c1b0a"
279
+ ],
280
+ "format": "uuid",
281
+ "type": "string"
282
+ }
283
+ },
284
+ "required": [
285
+ "query"
286
+ ],
287
+ "type": "object"
288
+ },
289
+ "annotations": {
290
+ "title": "Search transcriptions",
291
+ "readOnlyHint": true,
292
+ "destructiveHint": false,
293
+ "idempotentHint": true,
294
+ "openWorldHint": false
295
+ }
296
+ },
297
+ {
298
+ "name": "get_summary",
299
+ "title": "Get a summary",
300
+ "description": "Get the title and summary of a finished transcription, writing one the first time if it has none.\nUse: for an overview. Full text, or the whole text in another language: get_transcription. A specific question: search_transcriptions with mode meaning.\nBehaviour: free. Writing a first summary needs the write scope and keeps it for later calls; reading an existing one needs read only.\nReturns: state ready with title and summary (and lang when translated); pending while it is written or the transcription still runs (call again in a few seconds); not_requested for recordings under a minute; unavailable when the transcription failed.\nErrors: not_found; forbidden when a summary must be written and the key is read-only.",
301
+ "inputSchema": {
302
+ "additionalProperties": false,
303
+ "properties": {
304
+ "lang": {
305
+ "description": "Optional language code for the title and summary, honoured when the transcription was made with summary translations on; otherwise the original comes back without lang.",
306
+ "examples": [
307
+ "fr"
308
+ ],
309
+ "type": "string"
310
+ },
311
+ "transcriptionId": {
312
+ "description": "A transcriptionId from transcribe, list_transcriptions, search_transcriptions or start_broadcast.",
313
+ "examples": [
314
+ "0190f0e0-7c1a-7d3e-9b2a-5f4e3d2c1b0a"
315
+ ],
316
+ "format": "uuid",
317
+ "type": "string"
318
+ }
319
+ },
320
+ "required": [
321
+ "transcriptionId"
322
+ ],
323
+ "type": "object"
324
+ },
325
+ "annotations": {
326
+ "title": "Get a summary",
327
+ "readOnlyHint": false,
328
+ "destructiveHint": false,
329
+ "idempotentHint": true,
330
+ "openWorldHint": false
331
+ }
332
+ },
333
+ {
334
+ "name": "rename_speaker",
335
+ "title": "Rename a speaker",
336
+ "description": "Name a speaker in one transcription, so its transcript shows the name instead of \"Speaker N\".\nUse: labelling who spoke here. To recognise the person in later transcriptions too: save_speaker (it also names them here). Saved speakers: list_speakers.\nBehaviour: free; applies to that person in every recording of the transcription, and listeners following it live see it at once.\nReturns: transcriptionId, speaker, name (null when cleared).\nErrors: not_found for an unknown transcription or speaker number; forbidden for a read-only key.",
337
+ "inputSchema": {
338
+ "additionalProperties": false,
339
+ "properties": {
340
+ "name": {
341
+ "description": "The name to show instead of \"Speaker N\". An empty string clears it back to \"Speaker N\".",
342
+ "examples": [
343
+ "Maria Lopez"
344
+ ],
345
+ "maxLength": 80,
346
+ "type": "string"
347
+ },
348
+ "speaker": {
349
+ "description": "The N of \"Speaker N\" in get_transcription's text, or speakerNumber in its segments; it stays the same after naming.",
350
+ "examples": [
351
+ 2
352
+ ],
353
+ "minimum": 1,
354
+ "type": "integer"
355
+ },
356
+ "transcriptionId": {
357
+ "description": "A transcriptionId from transcribe, list_transcriptions, search_transcriptions or start_broadcast.",
358
+ "examples": [
359
+ "0190f0e0-7c1a-7d3e-9b2a-5f4e3d2c1b0a"
360
+ ],
361
+ "format": "uuid",
362
+ "type": "string"
363
+ }
364
+ },
365
+ "required": [
366
+ "transcriptionId",
367
+ "speaker",
368
+ "name"
369
+ ],
370
+ "type": "object"
371
+ },
372
+ "annotations": {
373
+ "title": "Rename a speaker",
374
+ "readOnlyHint": false,
375
+ "destructiveHint": false,
376
+ "idempotentHint": true,
377
+ "openWorldHint": false
378
+ }
379
+ },
380
+ {
381
+ "name": "delete_transcription",
382
+ "title": "Delete a transcription",
383
+ "description": "Move a transcription to the trash, with its recordings, transcript, translations and summary; its share links stop working.\nUse: to remove one. restore_transcription brings it back unchanged for 30 days; after that it is erased for good. Saved speakers stay (delete_speaker).\nBehaviour: a file or link still being transcribed is stopped first and keeps its text so far, billed only for what was transcribed; restoring does not resume it. A live one must be stopped first (stop_broadcast).\nReturns: transcriptionId, name, deleted, deletedAt, restorableUntil.\nErrors: invalid_state while recording live or when locked; not_found; forbidden for a read-only key.",
384
+ "inputSchema": {
385
+ "additionalProperties": false,
386
+ "properties": {
387
+ "transcriptionId": {
388
+ "description": "A transcriptionId from list_transcriptions (lifecycle archived lists the trash) or transcribe.",
389
+ "examples": [
390
+ "0190f0e0-7c1a-7d3e-9b2a-5f4e3d2c1b0a"
391
+ ],
392
+ "format": "uuid",
393
+ "type": "string"
394
+ }
395
+ },
396
+ "required": [
397
+ "transcriptionId"
398
+ ],
399
+ "type": "object"
400
+ },
401
+ "annotations": {
402
+ "title": "Delete a transcription",
403
+ "readOnlyHint": false,
404
+ "destructiveHint": true,
405
+ "idempotentHint": true,
406
+ "openWorldHint": false
407
+ }
408
+ },
409
+ {
410
+ "name": "restore_transcription",
411
+ "title": "Restore a transcription",
412
+ "description": "Bring a deleted transcription back from the trash, with its recordings, transcript, translations, summary and working share links.\nUse: undoing delete_transcription within 30 days; list_transcriptions with lifecycle archived lists the trash.\nBehaviour: free. A transcription stopped by its deletion stays stopped (transcribe it again to finish it).\nReturns: transcriptionId, name, restored true.\nErrors: not_found for an unknown id or one erased after its 30 days; forbidden for a read-only key.",
413
+ "inputSchema": {
414
+ "additionalProperties": false,
415
+ "properties": {
416
+ "transcriptionId": {
417
+ "description": "A transcriptionId from list_transcriptions (lifecycle archived lists the trash) or transcribe.",
418
+ "examples": [
419
+ "0190f0e0-7c1a-7d3e-9b2a-5f4e3d2c1b0a"
420
+ ],
421
+ "format": "uuid",
422
+ "type": "string"
423
+ }
424
+ },
425
+ "required": [
426
+ "transcriptionId"
427
+ ],
428
+ "type": "object"
429
+ },
430
+ "annotations": {
431
+ "title": "Restore a transcription",
432
+ "readOnlyHint": false,
433
+ "destructiveHint": false,
434
+ "idempotentHint": true,
435
+ "openWorldHint": false
436
+ }
437
+ },
438
+ {
439
+ "name": "set_cloned_voices",
440
+ "title": "Turn cloned voices on or off",
441
+ "description": "Turn cloned voices on or off for one transcription's spoken translations.\nOn: each speaker's translated words are read in that speaker's own cloned voice, never anyone else's; a speaker without a usable clone gets a stock voice. Off: stock voices for all. Voices are chosen by VocalStack; this is the only voice choice.\nConsent: turning on records, with your account and the time, that everyone speaking agreed, and releases the clones of this transcription waiting for it. Speakers who refused are never cloned.\nBehaviour: free; lines read aloud after the change use the new setting.\nReturns: transcriptionId, name, clonedVoices.\nErrors: invalid_options for on without consent, invalid_state for a deleted transcription, not_found, forbidden for a read-only key.",
442
+ "inputSchema": {
443
+ "additionalProperties": false,
444
+ "properties": {
445
+ "consent": {
446
+ "default": false,
447
+ "description": "Must be true to turn cloned voices on: confirms everyone speaking agreed to have their voice cloned. Recorded with your account and the time. Not needed to turn them off.",
448
+ "type": "boolean"
449
+ },
450
+ "on": {
451
+ "description": "true: read each speaker's translated words in their own cloned voice; false (every transcription's default): stock voices.",
452
+ "type": "boolean"
453
+ },
454
+ "transcriptionId": {
455
+ "description": "A transcriptionId from transcribe, list_transcriptions, search_transcriptions or start_broadcast.",
456
+ "examples": [
457
+ "0190f0e0-7c1a-7d3e-9b2a-5f4e3d2c1b0a"
458
+ ],
459
+ "format": "uuid",
460
+ "type": "string"
461
+ }
462
+ },
463
+ "required": [
464
+ "transcriptionId",
465
+ "on"
466
+ ],
467
+ "type": "object"
468
+ },
469
+ "annotations": {
470
+ "title": "Turn cloned voices on or off",
471
+ "readOnlyHint": false,
472
+ "destructiveHint": false,
473
+ "idempotentHint": true,
474
+ "openWorldHint": false
475
+ }
476
+ },
477
+ {
478
+ "name": "list_speakers",
479
+ "title": "List saved speakers",
480
+ "description": "List the speakers this account has saved: people recognised by name in every new transcription.\nUse: to see who is known or find a speakerId for delete_speaker. Save one: save_speaker. Names inside one transcription come from get_transcription and are set with rename_speaker, which saves no one.\nBehaviour: free; no paging.\nReturns: speakers, newest first, each with id, name, gender when known, cloned (their voice was cloned, with consent, to read translations of their own words) and, when cloned, voiceReady. None saved: an empty list.",
481
+ "inputSchema": {
482
+ "additionalProperties": false,
483
+ "properties": {},
484
+ "type": "object"
485
+ },
486
+ "annotations": {
487
+ "title": "List saved speakers",
488
+ "readOnlyHint": true,
489
+ "destructiveHint": false,
490
+ "idempotentHint": true,
491
+ "openWorldHint": false
492
+ }
493
+ },
494
+ {
495
+ "name": "save_speaker",
496
+ "title": "Save a speaker",
497
+ "description": "Save one speaker of a transcription to the account's saved speakers, so later transcriptions recognise and name them; their own speech in this transcription is the sample.\nUse: remembering a person across transcriptions. This transcription only: rename_speaker. Who is saved: list_speakers. Remove: delete_speaker.\nBehaviour: free; also names them in this transcription. Each call saves a new speaker, so check list_speakers first. Without consent only a voice fingerprint is kept; with it the voice is cloned too, recorded with your account and the time.\nReturns: speakerId, name, consent (granted or pending), cloned, voiceState (training, then ready) when cloned.\nErrors: not_found for an unknown transcription or speaker number; invalid_options for too little speech of them (the message says how much there is); forbidden for a read-only key.",
498
+ "inputSchema": {
499
+ "additionalProperties": false,
500
+ "properties": {
501
+ "consent": {
502
+ "default": false,
503
+ "description": "true only when this person agreed to have their voice cloned: it is then cloned too, from their own speech in this transcription, so translations of their words can be read in their voice. false: only their fingerprint is saved. Never set it without the person's agreement.",
504
+ "type": "boolean"
505
+ },
506
+ "name": {
507
+ "description": "The person's name, shown in this and later transcriptions.",
508
+ "examples": [
509
+ "Maria Lopez"
510
+ ],
511
+ "maxLength": 60,
512
+ "minLength": 1,
513
+ "type": "string"
514
+ },
515
+ "speaker": {
516
+ "description": "The N of \"Speaker N\" in get_transcription's text, or speakerNumber in its segments. Works while a live transcription is still running.",
517
+ "examples": [
518
+ 2
519
+ ],
520
+ "minimum": 1,
521
+ "type": "integer"
522
+ },
523
+ "transcriptionId": {
524
+ "description": "A transcriptionId from transcribe, list_transcriptions, search_transcriptions or start_broadcast.",
525
+ "examples": [
526
+ "0190f0e0-7c1a-7d3e-9b2a-5f4e3d2c1b0a"
527
+ ],
528
+ "format": "uuid",
529
+ "type": "string"
530
+ }
531
+ },
532
+ "required": [
533
+ "transcriptionId",
534
+ "speaker",
535
+ "name"
536
+ ],
537
+ "type": "object"
538
+ },
539
+ "annotations": {
540
+ "title": "Save a speaker",
541
+ "readOnlyHint": false,
542
+ "destructiveHint": false,
543
+ "idempotentHint": false,
544
+ "openWorldHint": false
545
+ }
546
+ },
547
+ {
548
+ "name": "delete_speaker",
549
+ "title": "Delete a saved speaker",
550
+ "description": "Delete a saved speaker for good: their name, fingerprint, stored sample and any cloned voice are erased at once and cannot be restored.\nUse: the person should no longer be recognised, or asked for their voice to be removed. A name inside one transcription: rename_speaker.\nBehaviour: free. Transcripts keep every line they spoke; lines labelled with their saved name go back to \"Speaker N\".\nReturns: speakerId, deleted true.\nErrors: not_found for an unknown or already deleted id; forbidden for a read-only key.",
551
+ "inputSchema": {
552
+ "additionalProperties": false,
553
+ "properties": {
554
+ "speakerId": {
555
+ "description": "A saved speaker's id: id in list_speakers, or speakerId from save_speaker.",
556
+ "examples": [
557
+ "0190f0e0-7c1a-7d3e-9b2a-5f4e3d2c1b0a"
558
+ ],
559
+ "format": "uuid",
560
+ "type": "string"
561
+ }
562
+ },
563
+ "required": [
564
+ "speakerId"
565
+ ],
566
+ "type": "object"
567
+ },
568
+ "annotations": {
569
+ "title": "Delete a saved speaker",
570
+ "readOnlyHint": false,
571
+ "destructiveHint": true,
572
+ "idempotentHint": true,
573
+ "openWorldHint": false
574
+ }
575
+ },
576
+ {
577
+ "name": "list_broadcasts",
578
+ "title": "List broadcasts",
579
+ "description": "List this account's broadcasts, or get one in detail. A broadcast is a public link where listeners follow a live transcription in their own language.\nUse: to find a broadcastId for start_broadcast or stop_broadcast, see which are on air, or (with broadcastId) a broadcast's recorded transcriptions and streams. Text: get_transcription; they also appear in list_transcriptions.\nBehaviour: free; most recently updated first, up to 500 per lifecycle, no paging.\nReturns: broadcasts (broadcastId, name, link, access, onAir, listenersNow, lastRecordedAt once it has recorded, archived for one in the trash; while on air transcriptionId, state, onAirSince). With broadcastId, one item that adds transcriptions (up to 20: transcriptionId, name, recording, lastRecordedAt) and streams (name, url). None: an empty list.\nErrors: not_found for an unknown broadcastId.",
580
+ "inputSchema": {
581
+ "additionalProperties": false,
582
+ "properties": {
583
+ "broadcastId": {
584
+ "description": "Optional: return only this broadcast, in detail (its recorded transcriptions and stream addresses too). lifecycle is then ignored.",
585
+ "examples": [
586
+ "0190f0e0-7c1a-7d3e-9b2a-5f4e3d2c1b0a"
587
+ ],
588
+ "format": "uuid",
589
+ "type": "string"
590
+ },
591
+ "lifecycle": {
592
+ "default": "active",
593
+ "description": "Which broadcasts: active those in use, archived those in the trash, all both.",
594
+ "enum": [
595
+ "active",
596
+ "archived",
597
+ "all"
598
+ ]
599
+ }
600
+ },
601
+ "type": "object"
602
+ },
603
+ "annotations": {
604
+ "title": "List broadcasts",
605
+ "readOnlyHint": true,
606
+ "destructiveHint": false,
607
+ "idempotentHint": true,
608
+ "openWorldHint": false
609
+ }
610
+ },
611
+ {
612
+ "name": "start_broadcast",
613
+ "title": "Start a broadcast",
614
+ "description": "Start live transcription and translation of a stream playing now, on a public link listeners open in their own language.\nUse: a live stream (YouTube live, internet radio, HLS, Icecast). A recording that has ended: transcribe. Reading it while it runs: get_transcription.\nList price: $0.60 an hour while it runs, billed per second; Pay as you go bills at least 5 minutes per transcription. Listeners and translations are free. It runs until stop_broadcast or the stream ends.\nTranscriptions running at once per account: 5 on Pay as you go, 50 on Premium; past that a start is refused with quota_exceeded.\nBehaviour: each call starts a new run unless it repeats an idempotencyKey; omitting broadcastId makes a new broadcast with a new link.\nReturns: broadcastId, name, link (anyone who opens it reads, or with spokenTranslations also hears, the stream in their language), transcriptionId, state.\nErrors: invalid_state (with link) when already on air, invalid_options for a non-http(s) address, insufficient_credit (with topUpUrl), quota_exceeded, not_found for an unknown broadcastId, forbidden for a key without the write and recordings:publish scopes.",
615
+ "inputSchema": {
616
+ "additionalProperties": false,
617
+ "properties": {
618
+ "broadcastId": {
619
+ "description": "Optional: run it on this existing broadcast (from list_broadcasts), so its listeners' link stays the same. Omit to make a new broadcast with a new link.",
620
+ "examples": [
621
+ "0190f0e0-7c1a-7d3e-9b2a-5f4e3d2c1b0a"
622
+ ],
623
+ "format": "uuid",
624
+ "type": "string"
625
+ },
626
+ "idempotencyKey": {
627
+ "description": "Any unique string of visible ASCII. Retrying with the same key and the same arguments within 24 h returns the first result (with replayed true) and nothing is started twice; the same key with other arguments is refused.",
628
+ "examples": [
629
+ "meeting-2026-11-19"
630
+ ],
631
+ "maxLength": 255,
632
+ "type": "string"
633
+ },
634
+ "name": {
635
+ "description": "Name of a new broadcast, shown to listeners. Defaults to the stream's host. Ignored with broadcastId.",
636
+ "examples": [
637
+ "Sunday service"
638
+ ],
639
+ "maxLength": 200,
640
+ "type": "string"
641
+ },
642
+ "spokenLanguage": {
643
+ "description": "Code of the language spoken, e.g. en, es. Omit to detect it.",
644
+ "examples": [
645
+ "en"
646
+ ],
647
+ "type": "string"
648
+ },
649
+ "spokenTranslations": {
650
+ "default": false,
651
+ "description": "Also read each translation aloud, so listeners can hear it in their language as well as read it. Listening stays free.",
652
+ "type": "boolean"
653
+ },
654
+ "streamUrl": {
655
+ "description": "http(s) address of a stream playing now: a YouTube live, an internet radio station, an HLS (.m3u8) or Icecast stream. A recording that has ended goes to transcribe instead.",
656
+ "examples": [
657
+ "https://www.youtube.com/watch?v=jfKfPfyJRdk",
658
+ "https://stream.example.org/radio.m3u8"
659
+ ],
660
+ "format": "uri",
661
+ "maxLength": 2048,
662
+ "type": "string"
663
+ }
664
+ },
665
+ "required": [
666
+ "streamUrl"
667
+ ],
668
+ "type": "object"
669
+ },
670
+ "annotations": {
671
+ "title": "Start a broadcast",
672
+ "readOnlyHint": false,
673
+ "destructiveHint": false,
674
+ "idempotentHint": false,
675
+ "openWorldHint": true
676
+ }
677
+ },
678
+ {
679
+ "name": "stop_broadcast",
680
+ "title": "Stop a broadcast",
681
+ "description": "Stop the live transcription a broadcast is running, which also stops its billing.\nUse: when the event is over. To remove the transcript afterwards: delete_transcription.\nBehaviour: free. The transcript is kept (get_transcription and get_summary work on it once state is done); the broadcastId and its link stay, ready for the next start_broadcast.\nReturns: the stopped transcription's status (transcriptionId, name, kind live, state finishing then done, startedAt, endedAt, durationS) with broadcastId and link.\nErrors: invalid_state when not on air, not_found for an unknown broadcastId, forbidden for a key without the recordings:publish scope.",
682
+ "inputSchema": {
683
+ "additionalProperties": false,
684
+ "properties": {
685
+ "broadcastId": {
686
+ "description": "A broadcastId from list_broadcasts or start_broadcast.",
687
+ "examples": [
688
+ "0190f0e0-7c1a-7d3e-9b2a-5f4e3d2c1b0a"
689
+ ],
690
+ "format": "uuid",
691
+ "type": "string"
692
+ }
693
+ },
694
+ "required": [
695
+ "broadcastId"
696
+ ],
697
+ "type": "object"
698
+ },
699
+ "annotations": {
700
+ "title": "Stop a broadcast",
701
+ "readOnlyHint": false,
702
+ "destructiveHint": false,
703
+ "idempotentHint": true,
704
+ "openWorldHint": false
705
+ }
706
+ }
707
+ ]
708
+ }