@tychenjiajun/pi-voice 0.7.0 → 0.8.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (33) hide show
  1. package/README.md +23 -1
  2. package/catalog/catalog.json +1501 -0
  3. package/catalog/recommendations.json +67 -27
  4. package/models/README.md +25 -0
  5. package/models/silero-vad-v6.2-F32.gguf +0 -0
  6. package/package.json +8 -6
  7. package/src/audio-constants.ts +2 -0
  8. package/src/audio.ts +2 -5
  9. package/src/catalog.ts +6 -2
  10. package/src/chinese.ts +17 -6
  11. package/src/chunking.ts +319 -0
  12. package/src/dictation-controller.ts +61 -18
  13. package/src/file-audio.ts +2 -2
  14. package/src/file-transcription.ts +6 -3
  15. package/src/language-detection-policy.ts +127 -0
  16. package/src/languages.ts +4 -3
  17. package/src/model-activation.ts +50 -4
  18. package/src/model-cells.ts +28 -6
  19. package/src/model-picker.ts +29 -12
  20. package/src/model-selection-controller.ts +14 -9
  21. package/src/models.ts +15 -1
  22. package/src/onboarding.ts +13 -1
  23. package/src/pcm-chunker.ts +14 -4
  24. package/src/recommendation-picker.ts +9 -7
  25. package/src/recommendations.ts +4 -2
  26. package/src/runtime.ts +9 -1
  27. package/src/settings-menu.ts +40 -9
  28. package/src/settings.ts +35 -3
  29. package/src/transcription-service.ts +205 -37
  30. package/src/transcription.ts +72 -24
  31. package/src/try-it.ts +11 -2
  32. package/src/vad.ts +301 -0
  33. package/src/catalog.generated.ts +0 -1305
package/src/catalog.generated.ts DELETED
@@ -1,1305 +0,0 @@
1
- // Generated from catalog/catalog.json by scripts/generate-catalog.mjs.
2
- // Do not edit by hand.
3
-
4
- export const CATALOG_MODELS_GENERATED = [
5
- {
6
- "id": "parakeet-unified-en-0.6b",
7
- "name": "Parakeet Unified EN 0.6B",
8
- "description": "Best all-around choice for English dictation",
9
- "repository": "handy-computer/parakeet-unified-en-0.6b-gguf",
10
- "revision": "7e948f21b7bdbac698d3318db9d350f1096f3b6c",
11
- "license": "other",
12
- "family": "parakeet",
13
- "parameters": "0.6B",
14
- "languages": [
15
- "en"
16
- ],
17
- "capabilities": {
18
- "streaming": true,
19
- "translate": false,
20
- "languageDetection": false
21
- },
22
- "quant": "Q8_0",
23
- "filename": "parakeet-unified-en-0.6b-Q8_0.gguf",
24
- "size": 731357568,
25
- "sha256": "4b50b6dd862bf6e346929aaf4f5eaacec003bfa3f56462d6c874b41ef2f38795"
26
- },
27
- {
28
- "id": "nemotron-3.5-asr-streaming-0.6b",
29
- "name": "Nemotron Streaming 3.5",
30
- "description": "Fast and accurate transcription in many European languages",
31
- "repository": "handy-computer/nemotron-3.5-asr-streaming-0.6b-gguf",
32
- "revision": "6d44e540bc31b0de1dbe174a3cea87f53a7f22fb",
33
- "license": "other",
34
- "family": "nemotron",
35
- "parameters": "0.6B",
36
- "languages": [
37
- "en-US",
38
- "en-GB",
39
- "es-US",
40
- "es-ES",
41
- "fr-FR",
42
- "fr-CA",
43
- "it-IT",
44
- "pt-BR",
45
- "pt-PT",
46
- "nl-NL",
47
- "de-DE",
48
- "tr-TR",
49
- "ru-RU",
50
- "ar-AR",
51
- "hi-IN",
52
- "ja-JP",
53
- "ko-KR",
54
- "vi-VN",
55
- "uk-UA",
56
- "pl-PL",
57
- "sv-SE",
58
- "cs-CZ",
59
- "nb-NO",
60
- "da-DK",
61
- "bg-BG",
62
- "fi-FI",
63
- "hr-HR",
64
- "sk-SK",
65
- "zh-CN",
66
- "hu-HU",
67
- "ro-RO",
68
- "et-EE"
69
- ],
70
- "capabilities": {
71
- "streaming": true,
72
- "translate": false,
73
- "languageDetection": true
74
- },
75
- "quant": "Q8_0",
76
- "filename": "nemotron-3.5-asr-streaming-0.6b-Q8_0.gguf",
77
- "size": 751094240,
78
- "sha256": "b94545b313b3223fda7b2857a52681da813935c2127643d1e9ff0c23d988089c"
79
- },
80
- {
81
- "id": "canary-180m-flash",
82
- "name": "Canary 180M Flash",
83
- "description": "Tiny, fast, and runs well on any hardware",
84
- "repository": "handy-computer/canary-180m-flash-gguf",
85
- "revision": "b147f9dc52b59f0998e410540a84727bd86457fd",
86
- "license": "cc-by-4.0",
87
- "family": "canary",
88
- "parameters": "180M",
89
- "languages": [
90
- "en",
91
- "de",
92
- "es",
93
- "fr"
94
- ],
95
- "capabilities": {
96
- "streaming": false,
97
- "translate": true,
98
- "languageDetection": false
99
- },
100
- "quant": "Q8_0",
101
- "filename": "canary-180m-flash-Q8_0.gguf",
102
- "size": 218447552,
103
- "sha256": "e13c7f5d0952b056a027cfffec13e3a3a134d1608babed24f983568f141e297c"
104
- },
105
- {
106
- "id": "cohere-transcribe-03-2026",
107
- "name": "Cohere Transcribe",
108
- "description": "Consistently strong in many languages, best for Greek and Polish",
109
- "repository": "handy-computer/cohere-transcribe-03-2026-gguf",
110
- "revision": "dfa4adebb64f3076b7b6b90b721275cc069cb421",
111
- "license": "apache-2.0",
112
- "family": "cohere",
113
- "parameters": "2.0B",
114
- "languages": [
115
- "en",
116
- "fr",
117
- "de",
118
- "es",
119
- "it",
120
- "pt",
121
- "nl",
122
- "pl",
123
- "el",
124
- "ar",
125
- "ja",
126
- "zh",
127
- "vi",
128
- "ko"
129
- ],
130
- "capabilities": {
131
- "streaming": false,
132
- "translate": false,
133
- "languageDetection": false
134
- },
135
- "quant": "Q8_0",
136
- "filename": "cohere-transcribe-03-2026-Q8_0.gguf",
137
- "size": 2410655232,
138
- "sha256": "931916663432fd895423a4291a8400221802b288967ca2d435fc5e3141c9e71e"
139
- },
140
- {
141
- "id": "whisper-medium",
142
- "name": "Whisper Medium",
143
- "description": "Broadest language support, but may run a bit slow",
144
- "repository": "handy-computer/whisper-medium-gguf",
145
- "revision": "ec78f06fded51aa82cde751678b78f76f78c8b7f",
146
- "license": "apache-2.0",
147
- "family": "whisper",
148
- "parameters": "764M",
149
- "languages": [
150
- "af",
151
- "am",
152
- "ar",
153
- "as",
154
- "az",
155
- "ba",
156
- "be",
157
- "bg",
158
- "bn",
159
- "bo",
160
- "br",
161
- "bs",
162
- "ca",
163
- "cs",
164
- "cy",
165
- "da",
166
- "de",
167
- "el",
168
- "en",
169
- "es",
170
- "et",
171
- "eu",
172
- "fa",
173
- "fi",
174
- "fo",
175
- "fr",
176
- "gl",
177
- "gu",
178
- "haw",
179
- "ha",
180
- "he",
181
- "hi",
182
- "hr",
183
- "ht",
184
- "hu",
185
- "hy",
186
- "id",
187
- "is",
188
- "it",
189
- "ja",
190
- "jw",
191
- "ka",
192
- "kk",
193
- "km",
194
- "kn",
195
- "ko",
196
- "la",
197
- "lb",
198
- "ln",
199
- "lo",
200
- "lt",
201
- "lv",
202
- "mg",
203
- "mi",
204
- "mk",
205
- "ml",
206
- "mn",
207
- "mr",
208
- "ms",
209
- "mt",
210
- "my",
211
- "ne",
212
- "nl",
213
- "nn",
214
- "no",
215
- "oc",
216
- "pa",
217
- "pl",
218
- "ps",
219
- "pt",
220
- "ro",
221
- "ru",
222
- "sa",
223
- "sd",
224
- "si",
225
- "sk",
226
- "sl",
227
- "sn",
228
- "so",
229
- "sq",
230
- "sr",
231
- "su",
232
- "sv",
233
- "sw",
234
- "ta",
235
- "te",
236
- "tg",
237
- "th",
238
- "tk",
239
- "tl",
240
- "tr",
241
- "tt",
242
- "uk",
243
- "ur",
244
- "uz",
245
- "vi",
246
- "yi",
247
- "yo",
248
- "zh"
249
- ],
250
- "capabilities": {
251
- "streaming": false,
252
- "translate": true,
253
- "languageDetection": true
254
- },
255
- "quant": "Q8_0",
256
- "filename": "whisper-medium-Q8_0.gguf",
257
- "size": 831538144,
258
- "sha256": "09e6a65e7de377aa5b10bae24608bc6f8ca2ed04b3993ef10d4a02bcd9a82adf"
259
- },
260
- {
261
- "id": "Voxtral-Mini-4B-Realtime-2602",
262
- "name": "Voxtral Mini 4B Realtime",
263
- "description": "Multilingual, very accurate in Italian and Spanish. Best on Apple Silicon or discrete GPUs",
264
- "repository": "handy-computer/Voxtral-Mini-4B-Realtime-2602-gguf",
265
- "revision": "b3e1c979e3775cbd0a49a65878a0ec7f06789ed7",
266
- "license": "apache-2.0",
267
- "family": "voxtral",
268
- "parameters": "4.8B",
269
- "languages": [
270
- "en",
271
- "fr",
272
- "es",
273
- "de",
274
- "ru",
275
- "zh",
276
- "ja",
277
- "it",
278
- "pt",
279
- "nl",
280
- "ar",
281
- "hi",
282
- "ko"
283
- ],
284
- "capabilities": {
285
- "streaming": true,
286
- "translate": false,
287
- "languageDetection": true
288
- },
289
- "quant": "Q8_0",
290
- "filename": "Voxtral-Mini-4B-Realtime-2602-Q8_0.gguf",
291
- "size": 4731791648,
292
- "sha256": "6fb22249463c1e7a600d920ad732e2dd0a231510e9db59fb8061d3d1d48d0c67"
293
- },
294
- {
295
- "id": "parakeet-tdt-0.6b-v3",
296
- "name": "Parakeet TDT 0.6B v3",
297
- "description": "Fast and accurate, great overall multilingual pick",
298
- "repository": "handy-computer/parakeet-tdt-0.6b-v3-gguf",
299
- "revision": "85ac09ea12fc4b1112fa76810059364bc6adc9de",
300
- "license": "cc-by-4.0",
301
- "family": "parakeet",
302
- "parameters": "0.6B",
303
- "languages": [
304
- "bg",
305
- "hr",
306
- "cs",
307
- "da",
308
- "nl",
309
- "en",
310
- "et",
311
- "fi",
312
- "fr",
313
- "de",
314
- "el",
315
- "hu",
316
- "it",
317
- "lv",
318
- "lt",
319
- "mt",
320
- "pl",
321
- "pt",
322
- "ro",
323
- "ru",
324
- "sk",
325
- "sl",
326
- "es",
327
- "sv",
328
- "uk"
329
- ],
330
- "capabilities": {
331
- "streaming": false,
332
- "translate": false,
333
- "languageDetection": true
334
- },
335
- "quant": "Q8_0",
336
- "filename": "parakeet-tdt-0.6b-v3-Q8_0.gguf",
337
- "size": 739508576,
338
- "sha256": "5859f77944efcd8eafa23a6350731960b2b55b2203df51f319665c807d802cc7"
339
- },
340
- {
341
- "id": "parakeet-tdt-0.6b-v2",
342
- "name": "Parakeet TDT 0.6B v2",
343
- "description": "Excellent English only model, runs on any hardware",
344
- "repository": "handy-computer/parakeet-tdt-0.6b-v2-gguf",
345
- "revision": "07cee0616125a08ef619729bb47f40ef747e4bc4",
346
- "license": "cc-by-4.0",
347
- "family": "parakeet",
348
- "parameters": "0.6B",
349
- "languages": [
350
- "en"
351
- ],
352
- "capabilities": {
353
- "streaming": false,
354
- "translate": false,
355
- "languageDetection": false
356
- },
357
- "quant": "Q8_0",
358
- "filename": "parakeet-tdt-0.6b-v2-Q8_0.gguf",
359
- "size": 729574912,
360
- "sha256": "f0d0e99cebb6d3b83f1f7069b82b5d3c2e39a54545b0da039cb4bafd9c4e5caa"
361
- },
362
- {
363
- "id": "Qwen3-ASR-0.6B",
364
- "name": "Qwen3-ASR 0.6B",
365
- "description": "Best for Chinese, Cantonese, and Asian languages with English",
366
- "repository": "handy-computer/Qwen3-ASR-0.6B-gguf",
367
- "revision": "e4e16599b900eb0cb36e524514756bb92eb092b7",
368
- "license": "apache-2.0",
369
- "family": "qwen3",
370
- "parameters": "782M",
371
- "languages": [
372
- "zh",
373
- "en",
374
- "yue",
375
- "ar",
376
- "de",
377
- "fr",
378
- "es",
379
- "pt",
380
- "id",
381
- "it",
382
- "ko",
383
- "ru",
384
- "th",
385
- "vi",
386
- "ja",
387
- "tr",
388
- "hi",
389
- "ms",
390
- "nl",
391
- "sv",
392
- "da",
393
- "fi",
394
- "pl",
395
- "cs",
396
- "fil",
397
- "fa",
398
- "el",
399
- "ro",
400
- "hu",
401
- "mk"
402
- ],
403
- "capabilities": {
404
- "streaming": false,
405
- "translate": false,
406
- "languageDetection": true
407
- },
408
- "quant": "Q8_0",
409
- "filename": "Qwen3-ASR-0.6B-Q8_0.gguf",
410
- "size": 850423456,
411
- "sha256": "f081b2d5e23bd669d92cc331d722a8a0681943b8e6f34b48996fd5c319b5acd8"
412
- },
413
- {
414
- "id": "Fun-ASR-MLT-Nano-2512",
415
- "name": "Fun-ASR Nano Multilingual",
416
- "description": "Best for Japanese, good at Thai, Indonesian, Malay",
417
- "repository": "handy-computer/Fun-ASR-MLT-Nano-2512-gguf",
418
- "revision": "0b8f9c7bc545a219658aeb1dd4eeaa55d1cf89f3",
419
- "license": "apache-2.0",
420
- "family": "fun",
421
- "parameters": "985M",
422
- "languages": [
423
- "zh",
424
- "en",
425
- "yue",
426
- "ja",
427
- "ko",
428
- "vi",
429
- "id",
430
- "th",
431
- "ms",
432
- "tl",
433
- "ar",
434
- "hi",
435
- "bg",
436
- "hr",
437
- "cs",
438
- "da",
439
- "nl",
440
- "et",
441
- "fi",
442
- "el",
443
- "hu",
444
- "ga",
445
- "lv",
446
- "lt",
447
- "mt",
448
- "pl",
449
- "pt",
450
- "ro",
451
- "sk",
452
- "sl",
453
- "sv"
454
- ],
455
- "capabilities": {
456
- "streaming": false,
457
- "translate": false,
458
- "languageDetection": false
459
- },
460
- "quant": "Q8_0",
461
- "filename": "Fun-ASR-MLT-Nano-2512-Q8_0.gguf",
462
- "size": 891271232,
463
- "sha256": "d12476d8d9f2baa0ebf738fa955fa05ed33a654f1567289033a810c45d9d9002"
464
- },
465
- {
466
- "id": "canary-1b-v2",
467
- "name": "Canary 1B v2",
468
- "description": "Most accurate small model for European languages",
469
- "repository": "handy-computer/canary-1b-v2-gguf",
470
- "revision": "58d13c2c0102229aad45f7e19a77ddc42b41dd9a",
471
- "license": "cc-by-4.0",
472
- "family": "canary",
473
- "parameters": "1B",
474
- "languages": [
475
- "bg",
476
- "hr",
477
- "cs",
478
- "da",
479
- "nl",
480
- "en",
481
- "et",
482
- "fi",
483
- "fr",
484
- "de",
485
- "el",
486
- "hu",
487
- "it",
488
- "lv",
489
- "lt",
490
- "mt",
491
- "pl",
492
- "pt",
493
- "ro",
494
- "sk",
495
- "sl",
496
- "es",
497
- "sv",
498
- "ru",
499
- "uk"
500
- ],
501
- "capabilities": {
502
- "streaming": false,
503
- "translate": true,
504
- "languageDetection": false
505
- },
506
- "quant": "Q8_0",
507
- "filename": "canary-1b-v2-Q8_0.gguf",
508
- "size": 1144290016,
509
- "sha256": "224f83d1bc487b3303b495a7d6874912fdece93de19d1a04b550829c30a5d289"
510
- },
511
- {
512
- "id": "canary-qwen-2.5b",
513
- "name": "Canary-Qwen 2.5B",
514
- "description": "Very accurate English. Slow without a GPU.",
515
- "repository": "handy-computer/canary-qwen-2.5b-gguf",
516
- "revision": "3370d4e2f28cc70eea79dfc9f2f43fb91eef3163",
517
- "license": "cc-by-4.0",
518
- "family": "canary",
519
- "parameters": "2.5B",
520
- "languages": [
521
- "en"
522
- ],
523
- "capabilities": {
524
- "streaming": false,
525
- "translate": false,
526
- "languageDetection": false
527
- },
528
- "quant": "Q8_0",
529
- "filename": "canary-qwen-2.5b-Q8_0.gguf",
530
- "size": 2797548928,
531
- "sha256": "d89aad1285d5bd5aa441c464d3a4cf37bd5474f70705408e71558d8627415b34"
532
- },
533
- {
534
- "id": "cohere-transcribe-arabic-07-2026",
535
- "name": "Cohere Transcribe Arabic",
536
- "description": "Best for Arabic and Arabic/English speakers",
537
- "repository": "handy-computer/cohere-transcribe-arabic-07-2026-gguf",
538
- "revision": "715cbe09ca9b60fc1e497db1e4478fa9fcf4ac21",
539
- "license": "apache-2.0",
540
- "family": "cohere",
541
- "parameters": "2.0B",
542
- "languages": [
543
- "en",
544
- "ar"
545
- ],
546
- "capabilities": {
547
- "streaming": false,
548
- "translate": false,
549
- "languageDetection": false
550
- },
551
- "quant": "Q8_0",
552
- "filename": "cohere-transcribe-arabic-07-2026-Q8_0.gguf",
553
- "size": 2410655136,
554
- "sha256": "910de5c9c57f9fd8a280e1701f9cd96f63768878c1ca9e4ecf23638a5e0fef16"
555
- },
556
- {
557
- "id": "Fun-ASR-Nano-2512",
558
- "name": "Fun-ASR Nano",
559
- "description": "Good for Mandarin, English, Japanese",
560
- "repository": "handy-computer/Fun-ASR-Nano-2512-gguf",
561
- "revision": "30360f003f99929c1e25c3f6222d02eccb04663c",
562
- "license": "apache-2.0",
563
- "family": "fun",
564
- "parameters": "985M",
565
- "languages": [
566
- "zh",
567
- "en",
568
- "ja"
569
- ],
570
- "capabilities": {
571
- "streaming": false,
572
- "translate": false,
573
- "languageDetection": false
574
- },
575
- "quant": "Q8_0",
576
- "filename": "Fun-ASR-Nano-2512-Q8_0.gguf",
577
- "size": 891270912,
578
- "sha256": "681caef6df15a2c0e153b40ca7fe4087fdf65751fa5e6fe605d8a75dff969e61"
579
- },
580
- {
581
- "id": "gigaam-v3-e2e-ctc",
582
- "name": "GigaAM v3 E2E-CTC",
583
- "description": "Russian only.",
584
- "repository": "handy-computer/gigaam-v3-e2e-ctc-gguf",
585
- "revision": "075dff81f843cf23d22b4ce943ffdc4dd8650cd7",
586
- "license": "mit",
587
- "family": "gigaam",
588
- "parameters": "221M",
589
- "languages": [
590
- "ru"
591
- ],
592
- "capabilities": {
593
- "streaming": false,
594
- "translate": false,
595
- "languageDetection": false
596
- },
597
- "quant": "Q8_0",
598
- "filename": "gigaam-v3-e2e-ctc-Q8_0.gguf",
599
- "size": 272151136,
600
- "sha256": "9ccce4750dc813a493d96ca15ee251712bedec15ac9a02fa3d2bd732f08ae5eb"
601
- },
602
- {
603
- "id": "gigaam-v3-e2e-rnnt",
604
- "name": "GigaAM v3 E2E-RNN-T",
605
- "description": "Best for Russian dictation. English may be unreliable.",
606
- "repository": "handy-computer/gigaam-v3-e2e-rnnt-gguf",
607
- "revision": "f719d70812344f4d0fb8c11c0887b190501a7465",
608
- "license": "mit",
609
- "family": "gigaam",
610
- "parameters": "223M",
611
- "languages": [
612
- "ru"
613
- ],
614
- "capabilities": {
615
- "streaming": false,
616
- "translate": false,
617
- "languageDetection": false
618
- },
619
- "quant": "Q8_0",
620
- "filename": "gigaam-v3-e2e-rnnt-Q8_0.gguf",
621
- "size": 273724832,
622
- "sha256": "78d63b47723b7f8d78c6113a6ef983b5a86e2a86f6c273e1f5cb6967b1c4467a"
623
- },
624
- {
625
- "id": "granite-speech-4.1-2b",
626
- "name": "Granite Speech 4.1 2B",
627
- "description": "Accurate with translation support. Best with GPU",
628
- "repository": "handy-computer/granite-speech-4.1-2b-gguf",
629
- "revision": "58e7710fd7039ded5a185668eef5f71ca5d9d919",
630
- "license": "apache-2.0",
631
- "family": "granite",
632
- "parameters": "2.3B",
633
- "languages": [
634
- "en",
635
- "fr",
636
- "de",
637
- "es",
638
- "pt",
639
- "ja"
640
- ],
641
- "capabilities": {
642
- "streaming": false,
643
- "translate": true,
644
- "languageDetection": false
645
- },
646
- "quant": "Q8_0",
647
- "filename": "granite-speech-4.1-2b-Q8_0.gguf",
648
- "size": 2559878848,
649
- "sha256": "8c0b2dce2861feb1dba91e9d59f2ab4a9c23ad808bff86465151576e096ec7a6"
650
- },
651
- {
652
- "id": "medasr",
653
- "name": "MedASR",
654
- "description": "For clinicians dictating notes. Poor on general speech",
655
- "repository": "handy-computer/medasr-gguf",
656
- "revision": "6f481df085bb50ae922cea918fb578e664237126",
657
- "license": "other",
658
- "family": "medasr",
659
- "parameters": "105M",
660
- "languages": [
661
- "en"
662
- ],
663
- "capabilities": {
664
- "streaming": false,
665
- "translate": false,
666
- "languageDetection": false
667
- },
668
- "quant": "Q8_0",
669
- "filename": "medasr-Q8_0.gguf",
670
- "size": 127712448,
671
- "sha256": "5a391a2154416b96241b829ac2c7eed8b64d197f00372512c9bbf63f4849c978"
672
- },
673
- {
674
- "id": "parakeet-tdt_ctc-110m",
675
- "name": "Parakeet TDT-CTC 110M",
676
- "description": "Tiny English model, virtually instant even on CPU",
677
- "repository": "handy-computer/parakeet-tdt_ctc-110m-gguf",
678
- "revision": "9d66d34f9e1594075c5dd72c90c0f4c321b29f21",
679
- "license": "cc-by-4.0",
680
- "family": "parakeet",
681
- "parameters": "110M",
682
- "languages": [
683
- "en"
684
- ],
685
- "capabilities": {
686
- "streaming": false,
687
- "translate": false,
688
- "languageDetection": false
689
- },
690
- "quant": "Q8_0",
691
- "filename": "parakeet-tdt_ctc-110m-Q8_0.gguf",
692
- "size": 135373280,
693
- "sha256": "7dd44c74a331d788a4e5f8b16913b3feb29ced22cf5613aad0e0f6cd30516296"
694
- },
695
- {
696
- "id": "multitalker-parakeet-streaming-0.6b-v1",
697
- "name": "Parakeet Multitalker EN",
698
- "description": "For overlapping speakers in meetings, not for solo dictation",
699
- "repository": "handy-computer/multitalker-parakeet-streaming-0.6b-v1-gguf",
700
- "revision": "0314bc0d2f4be9a0d79fa5eeb36ad1f3962cfcbb",
701
- "license": "other",
702
- "family": "parakeet",
703
- "parameters": "0.6B",
704
- "languages": [
705
- "en"
706
- ],
707
- "capabilities": {
708
- "streaming": true,
709
- "translate": false,
710
- "languageDetection": false
711
- },
712
- "quant": "Q8_0",
713
- "filename": "multitalker-parakeet-streaming-0.6b-v1-Q8_0.gguf",
714
- "size": 734123712,
715
- "sha256": "4748deea0222eb9057ebdac1679eebb970bbe882ee10a96ae34596ffe83615c7"
716
- },
717
- {
718
- "id": "Qwen3-ASR-1.7B",
719
- "name": "Qwen3-ASR 1.7B",
720
- "description": "Top accuracy across languages. Best on Apple Silicon or a GPU",
721
- "repository": "handy-computer/Qwen3-ASR-1.7B-gguf",
722
- "revision": "92282af1610a2db19d66f2bef1e260f5deca782d",
723
- "license": "apache-2.0",
724
- "family": "qwen3",
725
- "parameters": "2.0B",
726
- "languages": [
727
- "zh",
728
- "en",
729
- "yue",
730
- "ar",
731
- "de",
732
- "fr",
733
- "es",
734
- "pt",
735
- "id",
736
- "it",
737
- "ko",
738
- "ru",
739
- "th",
740
- "vi",
741
- "ja",
742
- "tr",
743
- "hi",
744
- "ms",
745
- "nl",
746
- "sv",
747
- "da",
748
- "fi",
749
- "pl",
750
- "cs",
751
- "fil",
752
- "fa",
753
- "el",
754
- "ro",
755
- "hu",
756
- "mk"
757
- ],
758
- "capabilities": {
759
- "streaming": false,
760
- "translate": false,
761
- "languageDetection": true
762
- },
763
- "quant": "Q8_0",
764
- "filename": "Qwen3-ASR-1.7B-Q8_0.gguf",
765
- "size": 2185030624,
766
- "sha256": "9a0d81792dfea2d5f278b8a63deb3ea6e02139ce42c2301f32ea19c4f77526b7"
767
- },
768
- {
769
- "id": "SenseVoiceSmall",
770
- "name": "SenseVoice Small",
771
- "description": "Instant Mandarin, Japanese, Korean on any machine. Weak Cantonese",
772
- "repository": "handy-computer/SenseVoiceSmall-gguf",
773
- "revision": "4a08b8e900b38a977e32eb08d5d0697d6e72ba04",
774
- "license": "other",
775
- "family": "sensevoice",
776
- "parameters": "234M",
777
- "languages": [
778
- "zh",
779
- "yue",
780
- "en",
781
- "ja",
782
- "ko"
783
- ],
784
- "capabilities": {
785
- "streaming": false,
786
- "translate": false,
787
- "languageDetection": true
788
- },
789
- "quant": "Q8_0",
790
- "filename": "SenseVoiceSmall-Q8_0.gguf",
791
- "size": 252684608,
792
- "sha256": "6c759ee4c9748c9b3f7a5a60ca74f0f7e685fb9d45d1378fce7cfd62f59adf29"
793
- },
794
- {
795
- "id": "Voxtral-Mini-3B-2507",
796
- "name": "Voxtral Mini 3B",
797
- "description": "Accurate with translation support. Best on Apple Silicon or a GPU.",
798
- "repository": "handy-computer/Voxtral-Mini-3B-2507-gguf",
799
- "revision": "5690205813042c07cbaa86d2a9dcc585fcd31304",
800
- "license": "apache-2.0",
801
- "family": "voxtral",
802
- "parameters": "4.7B",
803
- "languages": [
804
- "en",
805
- "fr",
806
- "de",
807
- "es",
808
- "it",
809
- "pt",
810
- "nl",
811
- "hi"
812
- ],
813
- "capabilities": {
814
- "streaming": false,
815
- "translate": true,
816
- "languageDetection": true
817
- },
818
- "quant": "Q8_0",
819
- "filename": "Voxtral-Mini-3B-2507-Q8_0.gguf",
820
- "size": 5000084128,
821
- "sha256": "06b4dd04e32953ccab0d417125bdd728ec031431596ed5e80123c3c66f29ddc2"
822
- },
823
- {
824
- "id": "whisper-small",
825
- "name": "Whisper Small",
826
- "description": "Wide language support in a relatively small model.",
827
- "repository": "handy-computer/whisper-small-gguf",
828
- "revision": "c0214bd34be9296695486f838e0142f900803159",
829
- "license": "apache-2.0",
830
- "family": "whisper",
831
- "parameters": "242M",
832
- "languages": [
833
- "af",
834
- "am",
835
- "ar",
836
- "as",
837
- "az",
838
- "ba",
839
- "be",
840
- "bg",
841
- "bn",
842
- "bo",
843
- "br",
844
- "bs",
845
- "ca",
846
- "cs",
847
- "cy",
848
- "da",
849
- "de",
850
- "el",
851
- "en",
852
- "es",
853
- "et",
854
- "eu",
855
- "fa",
856
- "fi",
857
- "fo",
858
- "fr",
859
- "gl",
860
- "gu",
861
- "haw",
862
- "ha",
863
- "he",
864
- "hi",
865
- "hr",
866
- "ht",
867
- "hu",
868
- "hy",
869
- "id",
870
- "is",
871
- "it",
872
- "ja",
873
- "jw",
874
- "ka",
875
- "kk",
876
- "km",
877
- "kn",
878
- "ko",
879
- "la",
880
- "lb",
881
- "ln",
882
- "lo",
883
- "lt",
884
- "lv",
885
- "mg",
886
- "mi",
887
- "mk",
888
- "ml",
889
- "mn",
890
- "mr",
891
- "ms",
892
- "mt",
893
- "my",
894
- "ne",
895
- "nl",
896
- "nn",
897
- "no",
898
- "oc",
899
- "pa",
900
- "pl",
901
- "ps",
902
- "pt",
903
- "ro",
904
- "ru",
905
- "sa",
906
- "sd",
907
- "si",
908
- "sk",
909
- "sl",
910
- "sn",
911
- "so",
912
- "sq",
913
- "sr",
914
- "su",
915
- "sv",
916
- "sw",
917
- "ta",
918
- "te",
919
- "tg",
920
- "th",
921
- "tk",
922
- "tl",
923
- "tr",
924
- "tt",
925
- "uk",
926
- "ur",
927
- "uz",
928
- "vi",
929
- "yi",
930
- "yo",
931
- "zh"
932
- ],
933
- "capabilities": {
934
- "streaming": false,
935
- "translate": true,
936
- "languageDetection": true
937
- },
938
- "quant": "Q8_0",
939
- "filename": "whisper-small-Q8_0.gguf",
940
- "size": 269751136,
941
- "sha256": "9b9c8811bbcc82a7766f0fb0925614bdacb0923b2cc630daeac17108b655b860"
942
- },
943
- {
944
- "id": "whisper-large-v3-turbo",
945
- "name": "Whisper Large v3 Turbo",
946
- "description": "Accurate in many languages. Slow without a GPU",
947
- "repository": "handy-computer/whisper-large-v3-turbo-gguf",
948
- "revision": "5eaf945c7978e564bae5b28a5b1639dd93c2bfb1",
949
- "license": "apache-2.0",
950
- "family": "whisper",
951
- "parameters": "809M",
952
- "languages": [
953
- "af",
954
- "am",
955
- "ar",
956
- "as",
957
- "az",
958
- "ba",
959
- "be",
960
- "bg",
961
- "bn",
962
- "bo",
963
- "br",
964
- "bs",
965
- "ca",
966
- "cs",
967
- "cy",
968
- "da",
969
- "de",
970
- "el",
971
- "en",
972
- "es",
973
- "et",
974
- "eu",
975
- "fa",
976
- "fi",
977
- "fo",
978
- "fr",
979
- "gl",
980
- "gu",
981
- "haw",
982
- "ha",
983
- "he",
984
- "hi",
985
- "hr",
986
- "ht",
987
- "hu",
988
- "hy",
989
- "id",
990
- "is",
991
- "it",
992
- "ja",
993
- "jw",
994
- "ka",
995
- "kk",
996
- "km",
997
- "kn",
998
- "ko",
999
- "la",
1000
- "lb",
1001
- "ln",
1002
- "lo",
1003
- "lt",
1004
- "lv",
1005
- "mg",
1006
- "mi",
1007
- "mk",
1008
- "ml",
1009
- "mn",
1010
- "mr",
1011
- "ms",
1012
- "mt",
1013
- "my",
1014
- "ne",
1015
- "nl",
1016
- "nn",
1017
- "no",
1018
- "oc",
1019
- "pa",
1020
- "pl",
1021
- "ps",
1022
- "pt",
1023
- "ro",
1024
- "ru",
1025
- "sa",
1026
- "sd",
1027
- "si",
1028
- "sk",
1029
- "sl",
1030
- "sn",
1031
- "so",
1032
- "sq",
1033
- "sr",
1034
- "su",
1035
- "sv",
1036
- "sw",
1037
- "ta",
1038
- "te",
1039
- "tg",
1040
- "th",
1041
- "tk",
1042
- "tl",
1043
- "tr",
1044
- "tt",
1045
- "uk",
1046
- "ur",
1047
- "uz",
1048
- "vi",
1049
- "yi",
1050
- "yo",
1051
- "yue",
1052
- "zh"
1053
- ],
1054
- "capabilities": {
1055
- "streaming": false,
1056
- "translate": false,
1057
- "languageDetection": true
1058
- },
1059
- "quant": "Q8_0",
1060
- "filename": "whisper-large-v3-turbo-Q8_0.gguf",
1061
- "size": 886381760,
1062
- "sha256": "b2e30cc286bc9f3aba4db9099fc7403543497c05ce7100d0d83091ddfd25a183"
1063
- },
1064
- {
1065
- "id": "Breeze-ASR-25",
1066
- "name": "Breeze-ASR-25",
1067
- "description": "Taiwan Mandarin with English mixed in. Traditional script, no punctuation",
1068
- "repository": "handy-computer/Breeze-ASR-25-gguf",
1069
- "revision": "1e5e8d7295110e1e14e305e9cf7411d82711beaa",
1070
- "license": "apache-2.0",
1071
- "family": "whisper",
1072
- "parameters": "1.5B",
1073
- "languages": [
1074
- "af",
1075
- "am",
1076
- "ar",
1077
- "as",
1078
- "az",
1079
- "ba",
1080
- "be",
1081
- "bg",
1082
- "bn",
1083
- "bo",
1084
- "br",
1085
- "bs",
1086
- "ca",
1087
- "cs",
1088
- "cy",
1089
- "da",
1090
- "de",
1091
- "el",
1092
- "en",
1093
- "es",
1094
- "et",
1095
- "eu",
1096
- "fa",
1097
- "fi",
1098
- "fo",
1099
- "fr",
1100
- "gl",
1101
- "gu",
1102
- "haw",
1103
- "ha",
1104
- "he",
1105
- "hi",
1106
- "hr",
1107
- "ht",
1108
- "hu",
1109
- "hy",
1110
- "id",
1111
- "is",
1112
- "it",
1113
- "ja",
1114
- "jw",
1115
- "ka",
1116
- "kk",
1117
- "km",
1118
- "kn",
1119
- "ko",
1120
- "la",
1121
- "lb",
1122
- "ln",
1123
- "lo",
1124
- "lt",
1125
- "lv",
1126
- "mg",
1127
- "mi",
1128
- "mk",
1129
- "ml",
1130
- "mn",
1131
- "mr",
1132
- "ms",
1133
- "mt",
1134
- "my",
1135
- "ne",
1136
- "nl",
1137
- "nn",
1138
- "no",
1139
- "oc",
1140
- "pa",
1141
- "pl",
1142
- "ps",
1143
- "pt",
1144
- "ro",
1145
- "ru",
1146
- "sa",
1147
- "sd",
1148
- "si",
1149
- "sk",
1150
- "sl",
1151
- "sn",
1152
- "so",
1153
- "sq",
1154
- "sr",
1155
- "su",
1156
- "sv",
1157
- "sw",
1158
- "ta",
1159
- "te",
1160
- "tg",
1161
- "th",
1162
- "tk",
1163
- "tl",
1164
- "tr",
1165
- "tt",
1166
- "uk",
1167
- "ur",
1168
- "uz",
1169
- "vi",
1170
- "yi",
1171
- "yo",
1172
- "zh"
1173
- ],
1174
- "capabilities": {
1175
- "streaming": false,
1176
- "translate": true,
1177
- "languageDetection": true
1178
- },
1179
- "quant": "Q8_0",
1180
- "filename": "Breeze-ASR-25-Q8_0.gguf",
1181
- "size": 1667964224,
1182
- "sha256": "1650c163cd1623d13b585368ffd87d7e669c5ff23014511e0fa91bb4ce994756"
1183
- },
1184
- {
1185
- "id": "whisper-large-v3",
1186
- "name": "Whisper Large v3",
1187
- "description": "Most languages at usable accuracy. Best on Apple Silicon or a GPU.",
1188
- "repository": "handy-computer/whisper-large-v3-gguf",
1189
- "revision": "e3e29bee6389c7da4a141406f07bb80ddac5337c",
1190
- "license": "apache-2.0",
1191
- "family": "whisper",
1192
- "parameters": "1.5B",
1193
- "languages": [
1194
- "af",
1195
- "am",
1196
- "ar",
1197
- "as",
1198
- "az",
1199
- "ba",
1200
- "be",
1201
- "bg",
1202
- "bn",
1203
- "bo",
1204
- "br",
1205
- "bs",
1206
- "ca",
1207
- "cs",
1208
- "cy",
1209
- "da",
1210
- "de",
1211
- "el",
1212
- "en",
1213
- "es",
1214
- "et",
1215
- "eu",
1216
- "fa",
1217
- "fi",
1218
- "fo",
1219
- "fr",
1220
- "gl",
1221
- "gu",
1222
- "haw",
1223
- "ha",
1224
- "he",
1225
- "hi",
1226
- "hr",
1227
- "ht",
1228
- "hu",
1229
- "hy",
1230
- "id",
1231
- "is",
1232
- "it",
1233
- "ja",
1234
- "jw",
1235
- "ka",
1236
- "kk",
1237
- "km",
1238
- "kn",
1239
- "ko",
1240
- "la",
1241
- "lb",
1242
- "ln",
1243
- "lo",
1244
- "lt",
1245
- "lv",
1246
- "mg",
1247
- "mi",
1248
- "mk",
1249
- "ml",
1250
- "mn",
1251
- "mr",
1252
- "ms",
1253
- "mt",
1254
- "my",
1255
- "ne",
1256
- "nl",
1257
- "nn",
1258
- "no",
1259
- "oc",
1260
- "pa",
1261
- "pl",
1262
- "ps",
1263
- "pt",
1264
- "ro",
1265
- "ru",
1266
- "sa",
1267
- "sd",
1268
- "si",
1269
- "sk",
1270
- "sl",
1271
- "sn",
1272
- "so",
1273
- "sq",
1274
- "sr",
1275
- "su",
1276
- "sv",
1277
- "sw",
1278
- "ta",
1279
- "te",
1280
- "tg",
1281
- "th",
1282
- "tk",
1283
- "tl",
1284
- "tr",
1285
- "tt",
1286
- "uk",
1287
- "ur",
1288
- "uz",
1289
- "vi",
1290
- "yi",
1291
- "yo",
1292
- "yue",
1293
- "zh"
1294
- ],
1295
- "capabilities": {
1296
- "streaming": false,
1297
- "translate": true,
1298
- "languageDetection": true
1299
- },
1300
- "quant": "Q8_0",
1301
- "filename": "whisper-large-v3-Q8_0.gguf",
1302
- "size": 1668741440,
1303
- "sha256": "2fa1a5f179f8a511a53e2108db270aa4af3ce08cd976af4180e2854666bb4ba3"
1304
- }
1305
- ] as const;