@tychenjiajun/pi-voice 0.7.0 → 0.8.1

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 +1,1501 @@
1
+ {
2
+ "version": 1,
3
+ "models": [
4
+ {
5
+ "id": "parakeet-unified-en-0.6b",
6
+ "name": "Parakeet Unified EN 0.6B",
7
+ "description": "Best all-around choice for English dictation",
8
+ "repository": "handy-computer/parakeet-unified-en-0.6b-gguf",
9
+ "revision": "7e948f21b7bdbac698d3318db9d350f1096f3b6c",
10
+ "license": "other",
11
+ "family": "parakeet",
12
+ "parameters": "0.6B",
13
+ "languages": [
14
+ "en"
15
+ ],
16
+ "capabilities": {
17
+ "streaming": true,
18
+ "translate": false,
19
+ "languageDetection": false
20
+ },
21
+ "quant": "Q8_0",
22
+ "filename": "parakeet-unified-en-0.6b-Q8_0.gguf",
23
+ "size": 731357568,
24
+ "sha256": "4b50b6dd862bf6e346929aaf4f5eaacec003bfa3f56462d6c874b41ef2f38795"
25
+ },
26
+ {
27
+ "id": "nemotron-3.5-asr-streaming-0.6b",
28
+ "name": "Nemotron Streaming 3.5",
29
+ "description": "Fast and accurate transcription in many European languages",
30
+ "repository": "handy-computer/nemotron-3.5-asr-streaming-0.6b-gguf",
31
+ "revision": "6d44e540bc31b0de1dbe174a3cea87f53a7f22fb",
32
+ "license": "other",
33
+ "family": "nemotron",
34
+ "parameters": "0.6B",
35
+ "languages": [
36
+ "en-US",
37
+ "en-GB",
38
+ "es-US",
39
+ "es-ES",
40
+ "fr-FR",
41
+ "fr-CA",
42
+ "it-IT",
43
+ "pt-BR",
44
+ "pt-PT",
45
+ "nl-NL",
46
+ "de-DE",
47
+ "tr-TR",
48
+ "ru-RU",
49
+ "ar-AR",
50
+ "hi-IN",
51
+ "ja-JP",
52
+ "ko-KR",
53
+ "vi-VN",
54
+ "uk-UA",
55
+ "pl-PL",
56
+ "sv-SE",
57
+ "cs-CZ",
58
+ "nb-NO",
59
+ "da-DK",
60
+ "bg-BG",
61
+ "fi-FI",
62
+ "hr-HR",
63
+ "sk-SK",
64
+ "zh-CN",
65
+ "hu-HU",
66
+ "ro-RO",
67
+ "et-EE"
68
+ ],
69
+ "capabilities": {
70
+ "streaming": true,
71
+ "translate": false,
72
+ "languageDetection": true
73
+ },
74
+ "quant": "Q8_0",
75
+ "filename": "nemotron-3.5-asr-streaming-0.6b-Q8_0.gguf",
76
+ "size": 751094240,
77
+ "sha256": "b94545b313b3223fda7b2857a52681da813935c2127643d1e9ff0c23d988089c"
78
+ },
79
+ {
80
+ "id": "canary-180m-flash",
81
+ "name": "Canary 180M Flash",
82
+ "description": "Tiny, fast, and runs well on any hardware",
83
+ "repository": "handy-computer/canary-180m-flash-gguf",
84
+ "revision": "b147f9dc52b59f0998e410540a84727bd86457fd",
85
+ "license": "cc-by-4.0",
86
+ "family": "canary",
87
+ "parameters": "180M",
88
+ "languages": [
89
+ "en",
90
+ "de",
91
+ "es",
92
+ "fr"
93
+ ],
94
+ "capabilities": {
95
+ "streaming": false,
96
+ "translate": true,
97
+ "languageDetection": false
98
+ },
99
+ "quant": "Q8_0",
100
+ "filename": "canary-180m-flash-Q8_0.gguf",
101
+ "size": 218447552,
102
+ "sha256": "e13c7f5d0952b056a027cfffec13e3a3a134d1608babed24f983568f141e297c"
103
+ },
104
+ {
105
+ "id": "cohere-transcribe-03-2026",
106
+ "name": "Cohere Transcribe",
107
+ "description": "Consistently strong in many languages, best for Greek and Polish",
108
+ "repository": "handy-computer/cohere-transcribe-03-2026-gguf",
109
+ "revision": "dfa4adebb64f3076b7b6b90b721275cc069cb421",
110
+ "license": "apache-2.0",
111
+ "family": "cohere",
112
+ "parameters": "2.0B",
113
+ "languages": [
114
+ "en",
115
+ "fr",
116
+ "de",
117
+ "es",
118
+ "it",
119
+ "pt",
120
+ "nl",
121
+ "pl",
122
+ "el",
123
+ "ar",
124
+ "ja",
125
+ "zh",
126
+ "vi",
127
+ "ko"
128
+ ],
129
+ "capabilities": {
130
+ "streaming": false,
131
+ "translate": false,
132
+ "languageDetection": false
133
+ },
134
+ "quant": "Q8_0",
135
+ "filename": "cohere-transcribe-03-2026-Q8_0.gguf",
136
+ "size": 2410655232,
137
+ "sha256": "931916663432fd895423a4291a8400221802b288967ca2d435fc5e3141c9e71e"
138
+ },
139
+ {
140
+ "id": "whisper-medium",
141
+ "name": "Whisper Medium",
142
+ "description": "Broadest language support, but may run a bit slow",
143
+ "repository": "handy-computer/whisper-medium-gguf",
144
+ "revision": "ec78f06fded51aa82cde751678b78f76f78c8b7f",
145
+ "license": "apache-2.0",
146
+ "family": "whisper",
147
+ "parameters": "764M",
148
+ "languages": [
149
+ "af",
150
+ "am",
151
+ "ar",
152
+ "as",
153
+ "az",
154
+ "ba",
155
+ "be",
156
+ "bg",
157
+ "bn",
158
+ "bo",
159
+ "br",
160
+ "bs",
161
+ "ca",
162
+ "cs",
163
+ "cy",
164
+ "da",
165
+ "de",
166
+ "el",
167
+ "en",
168
+ "es",
169
+ "et",
170
+ "eu",
171
+ "fa",
172
+ "fi",
173
+ "fo",
174
+ "fr",
175
+ "gl",
176
+ "gu",
177
+ "haw",
178
+ "ha",
179
+ "he",
180
+ "hi",
181
+ "hr",
182
+ "ht",
183
+ "hu",
184
+ "hy",
185
+ "id",
186
+ "is",
187
+ "it",
188
+ "ja",
189
+ "jw",
190
+ "ka",
191
+ "kk",
192
+ "km",
193
+ "kn",
194
+ "ko",
195
+ "la",
196
+ "lb",
197
+ "ln",
198
+ "lo",
199
+ "lt",
200
+ "lv",
201
+ "mg",
202
+ "mi",
203
+ "mk",
204
+ "ml",
205
+ "mn",
206
+ "mr",
207
+ "ms",
208
+ "mt",
209
+ "my",
210
+ "ne",
211
+ "nl",
212
+ "nn",
213
+ "no",
214
+ "oc",
215
+ "pa",
216
+ "pl",
217
+ "ps",
218
+ "pt",
219
+ "ro",
220
+ "ru",
221
+ "sa",
222
+ "sd",
223
+ "si",
224
+ "sk",
225
+ "sl",
226
+ "sn",
227
+ "so",
228
+ "sq",
229
+ "sr",
230
+ "su",
231
+ "sv",
232
+ "sw",
233
+ "ta",
234
+ "te",
235
+ "tg",
236
+ "th",
237
+ "tk",
238
+ "tl",
239
+ "tr",
240
+ "tt",
241
+ "uk",
242
+ "ur",
243
+ "uz",
244
+ "vi",
245
+ "yi",
246
+ "yo",
247
+ "zh"
248
+ ],
249
+ "capabilities": {
250
+ "streaming": false,
251
+ "translate": true,
252
+ "languageDetection": true
253
+ },
254
+ "quant": "Q8_0",
255
+ "filename": "whisper-medium-Q8_0.gguf",
256
+ "size": 831538144,
257
+ "sha256": "09e6a65e7de377aa5b10bae24608bc6f8ca2ed04b3993ef10d4a02bcd9a82adf"
258
+ },
259
+ {
260
+ "id": "Voxtral-Mini-4B-Realtime-2602",
261
+ "name": "Voxtral Mini 4B Realtime",
262
+ "description": "Multilingual, very accurate in Italian and Spanish. Best on Apple Silicon or discrete GPUs",
263
+ "repository": "handy-computer/Voxtral-Mini-4B-Realtime-2602-gguf",
264
+ "revision": "b3e1c979e3775cbd0a49a65878a0ec7f06789ed7",
265
+ "license": "apache-2.0",
266
+ "family": "voxtral",
267
+ "parameters": "4.8B",
268
+ "languages": [
269
+ "en",
270
+ "fr",
271
+ "es",
272
+ "de",
273
+ "ru",
274
+ "zh",
275
+ "ja",
276
+ "it",
277
+ "pt",
278
+ "nl",
279
+ "ar",
280
+ "hi",
281
+ "ko"
282
+ ],
283
+ "capabilities": {
284
+ "streaming": true,
285
+ "translate": false,
286
+ "languageDetection": true
287
+ },
288
+ "quant": "Q8_0",
289
+ "filename": "Voxtral-Mini-4B-Realtime-2602-Q8_0.gguf",
290
+ "size": 4731791648,
291
+ "sha256": "6fb22249463c1e7a600d920ad732e2dd0a231510e9db59fb8061d3d1d48d0c67"
292
+ },
293
+ {
294
+ "id": "parakeet-ultra",
295
+ "name": "Parakeet Ultra",
296
+ "description": "Best accuracy across 25 European languages",
297
+ "repository": "handy-computer/parakeet-ultra-gguf",
298
+ "revision": "39eeb55181f0d354fd934f06e92fd8d5037fed8e",
299
+ "license": "cc-by-4.0",
300
+ "family": "parakeet",
301
+ "parameters": "0.6B",
302
+ "languages": [
303
+ "en",
304
+ "de",
305
+ "fr",
306
+ "es",
307
+ "it",
308
+ "pt",
309
+ "ru",
310
+ "uk",
311
+ "hr",
312
+ "sl",
313
+ "lv",
314
+ "lt",
315
+ "et",
316
+ "fi",
317
+ "sv",
318
+ "da",
319
+ "nl",
320
+ "pl",
321
+ "cs",
322
+ "sk",
323
+ "hu",
324
+ "ro",
325
+ "bg",
326
+ "el",
327
+ "mt"
328
+ ],
329
+ "capabilities": {
330
+ "streaming": false,
331
+ "translate": false,
332
+ "languageDetection": true
333
+ },
334
+ "quant": "Q8_0",
335
+ "filename": "parakeet-ultra-Q8_0.gguf",
336
+ "size": 740363168,
337
+ "sha256": "007a59761e9258779f189df396b50d12b83bb5b79e66df4b955c230d2f2a0a59"
338
+ },
339
+ {
340
+ "id": "parakeet-tdt-0.6b-v3",
341
+ "name": "Parakeet TDT 0.6B v3",
342
+ "description": "Fast and accurate, great overall multilingual pick",
343
+ "repository": "handy-computer/parakeet-tdt-0.6b-v3-gguf",
344
+ "revision": "85ac09ea12fc4b1112fa76810059364bc6adc9de",
345
+ "license": "cc-by-4.0",
346
+ "family": "parakeet",
347
+ "parameters": "0.6B",
348
+ "languages": [
349
+ "bg",
350
+ "hr",
351
+ "cs",
352
+ "da",
353
+ "nl",
354
+ "en",
355
+ "et",
356
+ "fi",
357
+ "fr",
358
+ "de",
359
+ "el",
360
+ "hu",
361
+ "it",
362
+ "lv",
363
+ "lt",
364
+ "mt",
365
+ "pl",
366
+ "pt",
367
+ "ro",
368
+ "ru",
369
+ "sk",
370
+ "sl",
371
+ "es",
372
+ "sv",
373
+ "uk"
374
+ ],
375
+ "capabilities": {
376
+ "streaming": false,
377
+ "translate": false,
378
+ "languageDetection": true
379
+ },
380
+ "quant": "Q8_0",
381
+ "filename": "parakeet-tdt-0.6b-v3-Q8_0.gguf",
382
+ "size": 739508576,
383
+ "sha256": "5859f77944efcd8eafa23a6350731960b2b55b2203df51f319665c807d802cc7"
384
+ },
385
+ {
386
+ "id": "parakeet-tdt-0.6b-v2",
387
+ "name": "Parakeet TDT 0.6B v2",
388
+ "description": "Excellent English only model, runs on any hardware",
389
+ "repository": "handy-computer/parakeet-tdt-0.6b-v2-gguf",
390
+ "revision": "07cee0616125a08ef619729bb47f40ef747e4bc4",
391
+ "license": "cc-by-4.0",
392
+ "family": "parakeet",
393
+ "parameters": "0.6B",
394
+ "languages": [
395
+ "en"
396
+ ],
397
+ "capabilities": {
398
+ "streaming": false,
399
+ "translate": false,
400
+ "languageDetection": false
401
+ },
402
+ "quant": "Q8_0",
403
+ "filename": "parakeet-tdt-0.6b-v2-Q8_0.gguf",
404
+ "size": 729574912,
405
+ "sha256": "f0d0e99cebb6d3b83f1f7069b82b5d3c2e39a54545b0da039cb4bafd9c4e5caa"
406
+ },
407
+ {
408
+ "id": "Qwen3-ASR-0.6B",
409
+ "name": "Qwen3-ASR 0.6B",
410
+ "description": "Best for Chinese, Cantonese, and Asian languages with English",
411
+ "repository": "handy-computer/Qwen3-ASR-0.6B-gguf",
412
+ "revision": "e4e16599b900eb0cb36e524514756bb92eb092b7",
413
+ "license": "apache-2.0",
414
+ "family": "qwen3",
415
+ "parameters": "782M",
416
+ "languages": [
417
+ "zh",
418
+ "en",
419
+ "yue",
420
+ "ar",
421
+ "de",
422
+ "fr",
423
+ "es",
424
+ "pt",
425
+ "id",
426
+ "it",
427
+ "ko",
428
+ "ru",
429
+ "th",
430
+ "vi",
431
+ "ja",
432
+ "tr",
433
+ "hi",
434
+ "ms",
435
+ "nl",
436
+ "sv",
437
+ "da",
438
+ "fi",
439
+ "pl",
440
+ "cs",
441
+ "fil",
442
+ "fa",
443
+ "el",
444
+ "ro",
445
+ "hu",
446
+ "mk"
447
+ ],
448
+ "capabilities": {
449
+ "streaming": false,
450
+ "translate": false,
451
+ "languageDetection": true
452
+ },
453
+ "quant": "Q8_0",
454
+ "filename": "Qwen3-ASR-0.6B-Q8_0.gguf",
455
+ "size": 850423456,
456
+ "sha256": "f081b2d5e23bd669d92cc331d722a8a0681943b8e6f34b48996fd5c319b5acd8"
457
+ },
458
+ {
459
+ "id": "Fun-ASR-MLT-Nano-2512",
460
+ "name": "Fun-ASR Nano Multilingual",
461
+ "description": "Best for Japanese, good at Thai, Indonesian, Malay",
462
+ "repository": "handy-computer/Fun-ASR-MLT-Nano-2512-gguf",
463
+ "revision": "0b8f9c7bc545a219658aeb1dd4eeaa55d1cf89f3",
464
+ "license": "apache-2.0",
465
+ "family": "fun",
466
+ "parameters": "985M",
467
+ "languages": [
468
+ "zh",
469
+ "en",
470
+ "yue",
471
+ "ja",
472
+ "ko",
473
+ "vi",
474
+ "id",
475
+ "th",
476
+ "ms",
477
+ "tl",
478
+ "ar",
479
+ "hi",
480
+ "bg",
481
+ "hr",
482
+ "cs",
483
+ "da",
484
+ "nl",
485
+ "et",
486
+ "fi",
487
+ "el",
488
+ "hu",
489
+ "ga",
490
+ "lv",
491
+ "lt",
492
+ "mt",
493
+ "pl",
494
+ "pt",
495
+ "ro",
496
+ "sk",
497
+ "sl",
498
+ "sv"
499
+ ],
500
+ "capabilities": {
501
+ "streaming": false,
502
+ "translate": false,
503
+ "languageDetection": false
504
+ },
505
+ "quant": "Q8_0",
506
+ "filename": "Fun-ASR-MLT-Nano-2512-Q8_0.gguf",
507
+ "size": 891271232,
508
+ "sha256": "d12476d8d9f2baa0ebf738fa955fa05ed33a654f1567289033a810c45d9d9002"
509
+ },
510
+ {
511
+ "id": "canary-1b-v2",
512
+ "name": "Canary 1B v2",
513
+ "description": "Most accurate small model for European languages",
514
+ "repository": "handy-computer/canary-1b-v2-gguf",
515
+ "revision": "58d13c2c0102229aad45f7e19a77ddc42b41dd9a",
516
+ "license": "cc-by-4.0",
517
+ "family": "canary",
518
+ "parameters": "1B",
519
+ "languages": [
520
+ "bg",
521
+ "hr",
522
+ "cs",
523
+ "da",
524
+ "nl",
525
+ "en",
526
+ "et",
527
+ "fi",
528
+ "fr",
529
+ "de",
530
+ "el",
531
+ "hu",
532
+ "it",
533
+ "lv",
534
+ "lt",
535
+ "mt",
536
+ "pl",
537
+ "pt",
538
+ "ro",
539
+ "sk",
540
+ "sl",
541
+ "es",
542
+ "sv",
543
+ "ru",
544
+ "uk"
545
+ ],
546
+ "capabilities": {
547
+ "streaming": false,
548
+ "translate": true,
549
+ "languageDetection": false
550
+ },
551
+ "quant": "Q8_0",
552
+ "filename": "canary-1b-v2-Q8_0.gguf",
553
+ "size": 1144290016,
554
+ "sha256": "224f83d1bc487b3303b495a7d6874912fdece93de19d1a04b550829c30a5d289"
555
+ },
556
+ {
557
+ "id": "canary-qwen-2.5b",
558
+ "name": "Canary-Qwen 2.5B",
559
+ "description": "Very accurate English. Slow without a GPU.",
560
+ "repository": "handy-computer/canary-qwen-2.5b-gguf",
561
+ "revision": "3370d4e2f28cc70eea79dfc9f2f43fb91eef3163",
562
+ "license": "cc-by-4.0",
563
+ "family": "canary",
564
+ "parameters": "2.5B",
565
+ "languages": [
566
+ "en"
567
+ ],
568
+ "capabilities": {
569
+ "streaming": false,
570
+ "translate": false,
571
+ "languageDetection": false
572
+ },
573
+ "quant": "Q8_0",
574
+ "filename": "canary-qwen-2.5b-Q8_0.gguf",
575
+ "size": 2797548928,
576
+ "sha256": "d89aad1285d5bd5aa441c464d3a4cf37bd5474f70705408e71558d8627415b34"
577
+ },
578
+ {
579
+ "id": "cohere-transcribe-arabic-07-2026",
580
+ "name": "Cohere Transcribe Arabic",
581
+ "description": "Best for Arabic and Arabic/English speakers",
582
+ "repository": "handy-computer/cohere-transcribe-arabic-07-2026-gguf",
583
+ "revision": "715cbe09ca9b60fc1e497db1e4478fa9fcf4ac21",
584
+ "license": "apache-2.0",
585
+ "family": "cohere",
586
+ "parameters": "2.0B",
587
+ "languages": [
588
+ "en",
589
+ "ar"
590
+ ],
591
+ "capabilities": {
592
+ "streaming": false,
593
+ "translate": false,
594
+ "languageDetection": false
595
+ },
596
+ "quant": "Q8_0",
597
+ "filename": "cohere-transcribe-arabic-07-2026-Q8_0.gguf",
598
+ "size": 2410655136,
599
+ "sha256": "910de5c9c57f9fd8a280e1701f9cd96f63768878c1ca9e4ecf23638a5e0fef16"
600
+ },
601
+ {
602
+ "id": "Fun-ASR-Nano-2512",
603
+ "name": "Fun-ASR Nano",
604
+ "description": "Good for Mandarin, English, Japanese",
605
+ "repository": "handy-computer/Fun-ASR-Nano-2512-gguf",
606
+ "revision": "30360f003f99929c1e25c3f6222d02eccb04663c",
607
+ "license": "apache-2.0",
608
+ "family": "fun",
609
+ "parameters": "985M",
610
+ "languages": [
611
+ "zh",
612
+ "en",
613
+ "ja"
614
+ ],
615
+ "capabilities": {
616
+ "streaming": false,
617
+ "translate": false,
618
+ "languageDetection": false
619
+ },
620
+ "quant": "Q8_0",
621
+ "filename": "Fun-ASR-Nano-2512-Q8_0.gguf",
622
+ "size": 891270912,
623
+ "sha256": "681caef6df15a2c0e153b40ca7fe4087fdf65751fa5e6fe605d8a75dff969e61"
624
+ },
625
+ {
626
+ "id": "gigaam-v3-e2e-ctc",
627
+ "name": "GigaAM v3 E2E-CTC",
628
+ "description": "Russian only.",
629
+ "repository": "handy-computer/gigaam-v3-e2e-ctc-gguf",
630
+ "revision": "075dff81f843cf23d22b4ce943ffdc4dd8650cd7",
631
+ "license": "mit",
632
+ "family": "gigaam",
633
+ "parameters": "221M",
634
+ "languages": [
635
+ "ru"
636
+ ],
637
+ "capabilities": {
638
+ "streaming": false,
639
+ "translate": false,
640
+ "languageDetection": false
641
+ },
642
+ "quant": "Q8_0",
643
+ "filename": "gigaam-v3-e2e-ctc-Q8_0.gguf",
644
+ "size": 272151136,
645
+ "sha256": "9ccce4750dc813a493d96ca15ee251712bedec15ac9a02fa3d2bd732f08ae5eb"
646
+ },
647
+ {
648
+ "id": "gigaam-v3-e2e-rnnt",
649
+ "name": "GigaAM v3 E2E-RNN-T",
650
+ "description": "Best for Russian dictation. English may be unreliable.",
651
+ "repository": "handy-computer/gigaam-v3-e2e-rnnt-gguf",
652
+ "revision": "f719d70812344f4d0fb8c11c0887b190501a7465",
653
+ "license": "mit",
654
+ "family": "gigaam",
655
+ "parameters": "223M",
656
+ "languages": [
657
+ "ru"
658
+ ],
659
+ "capabilities": {
660
+ "streaming": false,
661
+ "translate": false,
662
+ "languageDetection": false
663
+ },
664
+ "quant": "Q8_0",
665
+ "filename": "gigaam-v3-e2e-rnnt-Q8_0.gguf",
666
+ "size": 273724832,
667
+ "sha256": "78d63b47723b7f8d78c6113a6ef983b5a86e2a86f6c273e1f5cb6967b1c4467a"
668
+ },
669
+ {
670
+ "id": "granite-speech-5.0-470m-turboctc",
671
+ "name": "Granite Speech 5.0 470M TurboCTC",
672
+ "description": "Extremely fast English transcription, best on GPU",
673
+ "repository": "handy-computer/granite-speech-5.0-470m-turboctc-gguf",
674
+ "revision": "2b0b44d0d6c8b94a2e4f04a2cf3ec6a5c054463b",
675
+ "license": "apache-2.0",
676
+ "family": "granite",
677
+ "parameters": "470M",
678
+ "languages": [
679
+ "en"
680
+ ],
681
+ "capabilities": {
682
+ "streaming": false,
683
+ "translate": false,
684
+ "languageDetection": false
685
+ },
686
+ "quant": "Q8_0",
687
+ "filename": "granite-speech-5.0-470m-turboctc-Q8_0.gguf",
688
+ "size": 505606496,
689
+ "sha256": "0408fe0b33be19af5be423d8ba6b43768608b6f065e1ac97b78c84b55c140738"
690
+ },
691
+ {
692
+ "id": "granite-speech-4.1-2b",
693
+ "name": "Granite Speech 4.1 2B",
694
+ "description": "Accurate with translation support. Best with GPU",
695
+ "repository": "handy-computer/granite-speech-4.1-2b-gguf",
696
+ "revision": "58e7710fd7039ded5a185668eef5f71ca5d9d919",
697
+ "license": "apache-2.0",
698
+ "family": "granite",
699
+ "parameters": "2.3B",
700
+ "languages": [
701
+ "en",
702
+ "fr",
703
+ "de",
704
+ "es",
705
+ "pt",
706
+ "ja"
707
+ ],
708
+ "capabilities": {
709
+ "streaming": false,
710
+ "translate": true,
711
+ "languageDetection": false
712
+ },
713
+ "quant": "Q8_0",
714
+ "filename": "granite-speech-4.1-2b-Q8_0.gguf",
715
+ "size": 2559878848,
716
+ "sha256": "8c0b2dce2861feb1dba91e9d59f2ab4a9c23ad808bff86465151576e096ec7a6"
717
+ },
718
+ {
719
+ "id": "medasr",
720
+ "name": "MedASR",
721
+ "description": "For clinicians dictating notes. Poor on general speech",
722
+ "repository": "handy-computer/medasr-gguf",
723
+ "revision": "6f481df085bb50ae922cea918fb578e664237126",
724
+ "license": "other",
725
+ "family": "medasr",
726
+ "parameters": "105M",
727
+ "languages": [
728
+ "en"
729
+ ],
730
+ "capabilities": {
731
+ "streaming": false,
732
+ "translate": false,
733
+ "languageDetection": false
734
+ },
735
+ "quant": "Q8_0",
736
+ "filename": "medasr-Q8_0.gguf",
737
+ "size": 127712448,
738
+ "sha256": "5a391a2154416b96241b829ac2c7eed8b64d197f00372512c9bbf63f4849c978"
739
+ },
740
+ {
741
+ "id": "parakeet-tdt_ctc-110m",
742
+ "name": "Parakeet TDT-CTC 110M",
743
+ "description": "Tiny English model, virtually instant even on CPU",
744
+ "repository": "handy-computer/parakeet-tdt_ctc-110m-gguf",
745
+ "revision": "9d66d34f9e1594075c5dd72c90c0f4c321b29f21",
746
+ "license": "cc-by-4.0",
747
+ "family": "parakeet",
748
+ "parameters": "110M",
749
+ "languages": [
750
+ "en"
751
+ ],
752
+ "capabilities": {
753
+ "streaming": false,
754
+ "translate": false,
755
+ "languageDetection": false
756
+ },
757
+ "quant": "Q8_0",
758
+ "filename": "parakeet-tdt_ctc-110m-Q8_0.gguf",
759
+ "size": 135373280,
760
+ "sha256": "7dd44c74a331d788a4e5f8b16913b3feb29ced22cf5613aad0e0f6cd30516296"
761
+ },
762
+ {
763
+ "id": "multitalker-parakeet-streaming-0.6b-v1",
764
+ "name": "Parakeet Multitalker EN",
765
+ "description": "For overlapping speakers in meetings, not for solo dictation",
766
+ "repository": "handy-computer/multitalker-parakeet-streaming-0.6b-v1-gguf",
767
+ "revision": "0314bc0d2f4be9a0d79fa5eeb36ad1f3962cfcbb",
768
+ "license": "other",
769
+ "family": "parakeet",
770
+ "parameters": "0.6B",
771
+ "languages": [
772
+ "en"
773
+ ],
774
+ "capabilities": {
775
+ "streaming": true,
776
+ "translate": false,
777
+ "languageDetection": false
778
+ },
779
+ "quant": "Q8_0",
780
+ "filename": "multitalker-parakeet-streaming-0.6b-v1-Q8_0.gguf",
781
+ "size": 734123712,
782
+ "sha256": "4748deea0222eb9057ebdac1679eebb970bbe882ee10a96ae34596ffe83615c7"
783
+ },
784
+ {
785
+ "id": "Qwen3-ASR-1.7B",
786
+ "name": "Qwen3-ASR 1.7B",
787
+ "description": "Top accuracy across languages. Best on Apple Silicon or a GPU",
788
+ "repository": "handy-computer/Qwen3-ASR-1.7B-gguf",
789
+ "revision": "92282af1610a2db19d66f2bef1e260f5deca782d",
790
+ "license": "apache-2.0",
791
+ "family": "qwen3",
792
+ "parameters": "2.0B",
793
+ "languages": [
794
+ "zh",
795
+ "en",
796
+ "yue",
797
+ "ar",
798
+ "de",
799
+ "fr",
800
+ "es",
801
+ "pt",
802
+ "id",
803
+ "it",
804
+ "ko",
805
+ "ru",
806
+ "th",
807
+ "vi",
808
+ "ja",
809
+ "tr",
810
+ "hi",
811
+ "ms",
812
+ "nl",
813
+ "sv",
814
+ "da",
815
+ "fi",
816
+ "pl",
817
+ "cs",
818
+ "fil",
819
+ "fa",
820
+ "el",
821
+ "ro",
822
+ "hu",
823
+ "mk"
824
+ ],
825
+ "capabilities": {
826
+ "streaming": false,
827
+ "translate": false,
828
+ "languageDetection": true
829
+ },
830
+ "quant": "Q8_0",
831
+ "filename": "Qwen3-ASR-1.7B-Q8_0.gguf",
832
+ "size": 2185030624,
833
+ "sha256": "9a0d81792dfea2d5f278b8a63deb3ea6e02139ce42c2301f32ea19c4f77526b7"
834
+ },
835
+ {
836
+ "id": "SenseVoiceSmall",
837
+ "name": "SenseVoice Small",
838
+ "description": "Instant Mandarin, Japanese, Korean on any machine. Weak Cantonese",
839
+ "repository": "handy-computer/SenseVoiceSmall-gguf",
840
+ "revision": "4a08b8e900b38a977e32eb08d5d0697d6e72ba04",
841
+ "license": "other",
842
+ "family": "sensevoice",
843
+ "parameters": "234M",
844
+ "languages": [
845
+ "zh",
846
+ "yue",
847
+ "en",
848
+ "ja",
849
+ "ko"
850
+ ],
851
+ "capabilities": {
852
+ "streaming": false,
853
+ "translate": false,
854
+ "languageDetection": true
855
+ },
856
+ "quant": "Q8_0",
857
+ "filename": "SenseVoiceSmall-Q8_0.gguf",
858
+ "size": 252684608,
859
+ "sha256": "6c759ee4c9748c9b3f7a5a60ca74f0f7e685fb9d45d1378fce7cfd62f59adf29"
860
+ },
861
+ {
862
+ "id": "Voxtral-Mini-3B-2507",
863
+ "name": "Voxtral Mini 3B",
864
+ "description": "Accurate with translation support. Best on Apple Silicon or a GPU.",
865
+ "repository": "handy-computer/Voxtral-Mini-3B-2507-gguf",
866
+ "revision": "5690205813042c07cbaa86d2a9dcc585fcd31304",
867
+ "license": "apache-2.0",
868
+ "family": "voxtral",
869
+ "parameters": "4.7B",
870
+ "languages": [
871
+ "en",
872
+ "fr",
873
+ "de",
874
+ "es",
875
+ "it",
876
+ "pt",
877
+ "nl",
878
+ "hi"
879
+ ],
880
+ "capabilities": {
881
+ "streaming": false,
882
+ "translate": true,
883
+ "languageDetection": true
884
+ },
885
+ "quant": "Q8_0",
886
+ "filename": "Voxtral-Mini-3B-2507-Q8_0.gguf",
887
+ "size": 5000084128,
888
+ "sha256": "06b4dd04e32953ccab0d417125bdd728ec031431596ed5e80123c3c66f29ddc2"
889
+ },
890
+ {
891
+ "id": "whisper-small",
892
+ "name": "Whisper Small",
893
+ "description": "Wide language support in a relatively small model.",
894
+ "repository": "handy-computer/whisper-small-gguf",
895
+ "revision": "c0214bd34be9296695486f838e0142f900803159",
896
+ "license": "apache-2.0",
897
+ "family": "whisper",
898
+ "parameters": "242M",
899
+ "languages": [
900
+ "af",
901
+ "am",
902
+ "ar",
903
+ "as",
904
+ "az",
905
+ "ba",
906
+ "be",
907
+ "bg",
908
+ "bn",
909
+ "bo",
910
+ "br",
911
+ "bs",
912
+ "ca",
913
+ "cs",
914
+ "cy",
915
+ "da",
916
+ "de",
917
+ "el",
918
+ "en",
919
+ "es",
920
+ "et",
921
+ "eu",
922
+ "fa",
923
+ "fi",
924
+ "fo",
925
+ "fr",
926
+ "gl",
927
+ "gu",
928
+ "haw",
929
+ "ha",
930
+ "he",
931
+ "hi",
932
+ "hr",
933
+ "ht",
934
+ "hu",
935
+ "hy",
936
+ "id",
937
+ "is",
938
+ "it",
939
+ "ja",
940
+ "jw",
941
+ "ka",
942
+ "kk",
943
+ "km",
944
+ "kn",
945
+ "ko",
946
+ "la",
947
+ "lb",
948
+ "ln",
949
+ "lo",
950
+ "lt",
951
+ "lv",
952
+ "mg",
953
+ "mi",
954
+ "mk",
955
+ "ml",
956
+ "mn",
957
+ "mr",
958
+ "ms",
959
+ "mt",
960
+ "my",
961
+ "ne",
962
+ "nl",
963
+ "nn",
964
+ "no",
965
+ "oc",
966
+ "pa",
967
+ "pl",
968
+ "ps",
969
+ "pt",
970
+ "ro",
971
+ "ru",
972
+ "sa",
973
+ "sd",
974
+ "si",
975
+ "sk",
976
+ "sl",
977
+ "sn",
978
+ "so",
979
+ "sq",
980
+ "sr",
981
+ "su",
982
+ "sv",
983
+ "sw",
984
+ "ta",
985
+ "te",
986
+ "tg",
987
+ "th",
988
+ "tk",
989
+ "tl",
990
+ "tr",
991
+ "tt",
992
+ "uk",
993
+ "ur",
994
+ "uz",
995
+ "vi",
996
+ "yi",
997
+ "yo",
998
+ "zh"
999
+ ],
1000
+ "capabilities": {
1001
+ "streaming": false,
1002
+ "translate": true,
1003
+ "languageDetection": true
1004
+ },
1005
+ "quant": "Q8_0",
1006
+ "filename": "whisper-small-Q8_0.gguf",
1007
+ "size": 269751136,
1008
+ "sha256": "9b9c8811bbcc82a7766f0fb0925614bdacb0923b2cc630daeac17108b655b860"
1009
+ },
1010
+ {
1011
+ "id": "whisper-large-v3-turbo",
1012
+ "name": "Whisper Large v3 Turbo",
1013
+ "description": "Accurate in many languages. Slow without a GPU",
1014
+ "repository": "handy-computer/whisper-large-v3-turbo-gguf",
1015
+ "revision": "5eaf945c7978e564bae5b28a5b1639dd93c2bfb1",
1016
+ "license": "apache-2.0",
1017
+ "family": "whisper",
1018
+ "parameters": "809M",
1019
+ "languages": [
1020
+ "af",
1021
+ "am",
1022
+ "ar",
1023
+ "as",
1024
+ "az",
1025
+ "ba",
1026
+ "be",
1027
+ "bg",
1028
+ "bn",
1029
+ "bo",
1030
+ "br",
1031
+ "bs",
1032
+ "ca",
1033
+ "cs",
1034
+ "cy",
1035
+ "da",
1036
+ "de",
1037
+ "el",
1038
+ "en",
1039
+ "es",
1040
+ "et",
1041
+ "eu",
1042
+ "fa",
1043
+ "fi",
1044
+ "fo",
1045
+ "fr",
1046
+ "gl",
1047
+ "gu",
1048
+ "haw",
1049
+ "ha",
1050
+ "he",
1051
+ "hi",
1052
+ "hr",
1053
+ "ht",
1054
+ "hu",
1055
+ "hy",
1056
+ "id",
1057
+ "is",
1058
+ "it",
1059
+ "ja",
1060
+ "jw",
1061
+ "ka",
1062
+ "kk",
1063
+ "km",
1064
+ "kn",
1065
+ "ko",
1066
+ "la",
1067
+ "lb",
1068
+ "ln",
1069
+ "lo",
1070
+ "lt",
1071
+ "lv",
1072
+ "mg",
1073
+ "mi",
1074
+ "mk",
1075
+ "ml",
1076
+ "mn",
1077
+ "mr",
1078
+ "ms",
1079
+ "mt",
1080
+ "my",
1081
+ "ne",
1082
+ "nl",
1083
+ "nn",
1084
+ "no",
1085
+ "oc",
1086
+ "pa",
1087
+ "pl",
1088
+ "ps",
1089
+ "pt",
1090
+ "ro",
1091
+ "ru",
1092
+ "sa",
1093
+ "sd",
1094
+ "si",
1095
+ "sk",
1096
+ "sl",
1097
+ "sn",
1098
+ "so",
1099
+ "sq",
1100
+ "sr",
1101
+ "su",
1102
+ "sv",
1103
+ "sw",
1104
+ "ta",
1105
+ "te",
1106
+ "tg",
1107
+ "th",
1108
+ "tk",
1109
+ "tl",
1110
+ "tr",
1111
+ "tt",
1112
+ "uk",
1113
+ "ur",
1114
+ "uz",
1115
+ "vi",
1116
+ "yi",
1117
+ "yo",
1118
+ "yue",
1119
+ "zh"
1120
+ ],
1121
+ "capabilities": {
1122
+ "streaming": false,
1123
+ "translate": false,
1124
+ "languageDetection": true
1125
+ },
1126
+ "quant": "Q8_0",
1127
+ "filename": "whisper-large-v3-turbo-Q8_0.gguf",
1128
+ "size": 886381760,
1129
+ "sha256": "b2e30cc286bc9f3aba4db9099fc7403543497c05ce7100d0d83091ddfd25a183"
1130
+ },
1131
+ {
1132
+ "id": "Breeze-ASR-25",
1133
+ "name": "Breeze-ASR-25",
1134
+ "description": "Taiwan Mandarin with English mixed in. Traditional script, no punctuation",
1135
+ "repository": "handy-computer/Breeze-ASR-25-gguf",
1136
+ "revision": "1e5e8d7295110e1e14e305e9cf7411d82711beaa",
1137
+ "license": "apache-2.0",
1138
+ "family": "whisper",
1139
+ "parameters": "1.5B",
1140
+ "languages": [
1141
+ "af",
1142
+ "am",
1143
+ "ar",
1144
+ "as",
1145
+ "az",
1146
+ "ba",
1147
+ "be",
1148
+ "bg",
1149
+ "bn",
1150
+ "bo",
1151
+ "br",
1152
+ "bs",
1153
+ "ca",
1154
+ "cs",
1155
+ "cy",
1156
+ "da",
1157
+ "de",
1158
+ "el",
1159
+ "en",
1160
+ "es",
1161
+ "et",
1162
+ "eu",
1163
+ "fa",
1164
+ "fi",
1165
+ "fo",
1166
+ "fr",
1167
+ "gl",
1168
+ "gu",
1169
+ "haw",
1170
+ "ha",
1171
+ "he",
1172
+ "hi",
1173
+ "hr",
1174
+ "ht",
1175
+ "hu",
1176
+ "hy",
1177
+ "id",
1178
+ "is",
1179
+ "it",
1180
+ "ja",
1181
+ "jw",
1182
+ "ka",
1183
+ "kk",
1184
+ "km",
1185
+ "kn",
1186
+ "ko",
1187
+ "la",
1188
+ "lb",
1189
+ "ln",
1190
+ "lo",
1191
+ "lt",
1192
+ "lv",
1193
+ "mg",
1194
+ "mi",
1195
+ "mk",
1196
+ "ml",
1197
+ "mn",
1198
+ "mr",
1199
+ "ms",
1200
+ "mt",
1201
+ "my",
1202
+ "ne",
1203
+ "nl",
1204
+ "nn",
1205
+ "no",
1206
+ "oc",
1207
+ "pa",
1208
+ "pl",
1209
+ "ps",
1210
+ "pt",
1211
+ "ro",
1212
+ "ru",
1213
+ "sa",
1214
+ "sd",
1215
+ "si",
1216
+ "sk",
1217
+ "sl",
1218
+ "sn",
1219
+ "so",
1220
+ "sq",
1221
+ "sr",
1222
+ "su",
1223
+ "sv",
1224
+ "sw",
1225
+ "ta",
1226
+ "te",
1227
+ "tg",
1228
+ "th",
1229
+ "tk",
1230
+ "tl",
1231
+ "tr",
1232
+ "tt",
1233
+ "uk",
1234
+ "ur",
1235
+ "uz",
1236
+ "vi",
1237
+ "yi",
1238
+ "yo",
1239
+ "zh"
1240
+ ],
1241
+ "capabilities": {
1242
+ "streaming": false,
1243
+ "translate": true,
1244
+ "languageDetection": true
1245
+ },
1246
+ "quant": "Q8_0",
1247
+ "filename": "Breeze-ASR-25-Q8_0.gguf",
1248
+ "size": 1667964224,
1249
+ "sha256": "1650c163cd1623d13b585368ffd87d7e669c5ff23014511e0fa91bb4ce994756"
1250
+ },
1251
+ {
1252
+ "id": "whisper-large-v3",
1253
+ "name": "Whisper Large v3",
1254
+ "description": "Most languages at usable accuracy. Best on Apple Silicon or a GPU.",
1255
+ "repository": "handy-computer/whisper-large-v3-gguf",
1256
+ "revision": "e3e29bee6389c7da4a141406f07bb80ddac5337c",
1257
+ "license": "apache-2.0",
1258
+ "family": "whisper",
1259
+ "parameters": "1.5B",
1260
+ "languages": [
1261
+ "af",
1262
+ "am",
1263
+ "ar",
1264
+ "as",
1265
+ "az",
1266
+ "ba",
1267
+ "be",
1268
+ "bg",
1269
+ "bn",
1270
+ "bo",
1271
+ "br",
1272
+ "bs",
1273
+ "ca",
1274
+ "cs",
1275
+ "cy",
1276
+ "da",
1277
+ "de",
1278
+ "el",
1279
+ "en",
1280
+ "es",
1281
+ "et",
1282
+ "eu",
1283
+ "fa",
1284
+ "fi",
1285
+ "fo",
1286
+ "fr",
1287
+ "gl",
1288
+ "gu",
1289
+ "haw",
1290
+ "ha",
1291
+ "he",
1292
+ "hi",
1293
+ "hr",
1294
+ "ht",
1295
+ "hu",
1296
+ "hy",
1297
+ "id",
1298
+ "is",
1299
+ "it",
1300
+ "ja",
1301
+ "jw",
1302
+ "ka",
1303
+ "kk",
1304
+ "km",
1305
+ "kn",
1306
+ "ko",
1307
+ "la",
1308
+ "lb",
1309
+ "ln",
1310
+ "lo",
1311
+ "lt",
1312
+ "lv",
1313
+ "mg",
1314
+ "mi",
1315
+ "mk",
1316
+ "ml",
1317
+ "mn",
1318
+ "mr",
1319
+ "ms",
1320
+ "mt",
1321
+ "my",
1322
+ "ne",
1323
+ "nl",
1324
+ "nn",
1325
+ "no",
1326
+ "oc",
1327
+ "pa",
1328
+ "pl",
1329
+ "ps",
1330
+ "pt",
1331
+ "ro",
1332
+ "ru",
1333
+ "sa",
1334
+ "sd",
1335
+ "si",
1336
+ "sk",
1337
+ "sl",
1338
+ "sn",
1339
+ "so",
1340
+ "sq",
1341
+ "sr",
1342
+ "su",
1343
+ "sv",
1344
+ "sw",
1345
+ "ta",
1346
+ "te",
1347
+ "tg",
1348
+ "th",
1349
+ "tk",
1350
+ "tl",
1351
+ "tr",
1352
+ "tt",
1353
+ "uk",
1354
+ "ur",
1355
+ "uz",
1356
+ "vi",
1357
+ "yi",
1358
+ "yo",
1359
+ "yue",
1360
+ "zh"
1361
+ ],
1362
+ "capabilities": {
1363
+ "streaming": false,
1364
+ "translate": true,
1365
+ "languageDetection": true
1366
+ },
1367
+ "quant": "Q8_0",
1368
+ "filename": "whisper-large-v3-Q8_0.gguf",
1369
+ "size": 1668741440,
1370
+ "sha256": "2fa1a5f179f8a511a53e2108db270aa4af3ce08cd976af4180e2854666bb4ba3"
1371
+ }
1372
+ ],
1373
+ "languageDetector": {
1374
+ "id": "voxlingua107-ecapa-q8",
1375
+ "name": "VoxLingua107 language detector",
1376
+ "description": "Shared spoken-language detector for multilingual transcription models",
1377
+ "repository": "handy-computer/lang-id-voxlingua107-ecapa-gguf",
1378
+ "revision": "41cf26039e317c6b8883bf0a61b0785dc663fc26",
1379
+ "license": "apache-2.0",
1380
+ "family": "ECAPA-TDNN",
1381
+ "parameters": null,
1382
+ "languages": [
1383
+ "ab",
1384
+ "af",
1385
+ "am",
1386
+ "ar",
1387
+ "as",
1388
+ "az",
1389
+ "ba",
1390
+ "be",
1391
+ "bg",
1392
+ "bn",
1393
+ "bo",
1394
+ "br",
1395
+ "bs",
1396
+ "ca",
1397
+ "ceb",
1398
+ "cs",
1399
+ "cy",
1400
+ "da",
1401
+ "de",
1402
+ "el",
1403
+ "en",
1404
+ "eo",
1405
+ "es",
1406
+ "et",
1407
+ "eu",
1408
+ "fa",
1409
+ "fi",
1410
+ "fo",
1411
+ "fr",
1412
+ "gl",
1413
+ "gn",
1414
+ "gu",
1415
+ "gv",
1416
+ "ha",
1417
+ "haw",
1418
+ "hi",
1419
+ "hr",
1420
+ "ht",
1421
+ "hu",
1422
+ "hy",
1423
+ "ia",
1424
+ "id",
1425
+ "is",
1426
+ "it",
1427
+ "iw",
1428
+ "ja",
1429
+ "jw",
1430
+ "ka",
1431
+ "kk",
1432
+ "km",
1433
+ "kn",
1434
+ "ko",
1435
+ "la",
1436
+ "lb",
1437
+ "ln",
1438
+ "lo",
1439
+ "lt",
1440
+ "lv",
1441
+ "mg",
1442
+ "mi",
1443
+ "mk",
1444
+ "ml",
1445
+ "mn",
1446
+ "mr",
1447
+ "ms",
1448
+ "mt",
1449
+ "my",
1450
+ "ne",
1451
+ "nl",
1452
+ "nn",
1453
+ "no",
1454
+ "oc",
1455
+ "pa",
1456
+ "pl",
1457
+ "ps",
1458
+ "pt",
1459
+ "ro",
1460
+ "ru",
1461
+ "sa",
1462
+ "sco",
1463
+ "sd",
1464
+ "si",
1465
+ "sk",
1466
+ "sl",
1467
+ "sn",
1468
+ "so",
1469
+ "sq",
1470
+ "sr",
1471
+ "su",
1472
+ "sv",
1473
+ "sw",
1474
+ "ta",
1475
+ "te",
1476
+ "tg",
1477
+ "th",
1478
+ "tk",
1479
+ "tl",
1480
+ "tr",
1481
+ "tt",
1482
+ "uk",
1483
+ "ur",
1484
+ "uz",
1485
+ "vi",
1486
+ "war",
1487
+ "yi",
1488
+ "yo",
1489
+ "zh"
1490
+ ],
1491
+ "capabilities": {
1492
+ "streaming": false,
1493
+ "translate": false,
1494
+ "languageDetection": true
1495
+ },
1496
+ "quant": "Q8_0",
1497
+ "filename": "lang-id-voxlingua107-ecapa-Q8_0.gguf",
1498
+ "size": 26693632,
1499
+ "sha256": "063e6cb0af3d53085fc7d68356dfde5ecee2ffa7bc79855f7e4ff930081bfe3a"
1500
+ }
1501
+ }