Hugging Face Hub · seven audio pipeline tags

The State of Open Source Audio

58,722 open audio models
20K added in the last 12 months
1.9× the 12 months before that
7 tasks, from 36.6K models to 279

as of 2026-10-25

What open audio is made of

Every public model repo carrying one of the seven audio tags. Pick a field to follow it through the rest of the page.

New models per month, by field

Stacked, so the column height is all of open audio and each band is one field. ASR sets the scale; the fields moving fastest are underneath it.

01,0002,0003,0004,0002023202420252026Apr 2022ASR249TTS18Classification10Text-to-audio0Audio-to-audio8Audio-text-to-text0VAD2all audio287May 2022ASR350TTS11Classification13Text-to-audio0Audio-to-audio0Audio-text-to-text0VAD0all audio374Jun 2022ASR235TTS6Classification16Text-to-audio0Audio-to-audio5Audio-text-to-text0VAD0all audio262Jul 2022ASR1,230TTS1Classification16Text-to-audio0Audio-to-audio2Audio-text-to-text0VAD0all audio1,249Aug 2022ASR149TTS4Classification16Text-to-audio0Audio-to-audio0Audio-text-to-text0VAD0all audio169Sep 2022ASR169TTS9Classification14Text-to-audio0Audio-to-audio2Audio-text-to-text0VAD0all audio194Oct 2022ASR171TTS8Classification23Text-to-audio0Audio-to-audio50Audio-text-to-text0VAD2all audio254Nov 2022ASR407TTS6Classification21Text-to-audio0Audio-to-audio0Audio-text-to-text0VAD2all audio436Dec 2022ASR1,124TTS4Classification11Text-to-audio12Audio-to-audio2Audio-text-to-text0VAD0all audio1,153Jan 2023ASR295TTS9Classification7Text-to-audio1Audio-to-audio9Audio-text-to-text0VAD0all audio321Feb 2023ASR427TTS7Classification40Text-to-audio0Audio-to-audio3Audio-text-to-text0VAD0all audio477Mar 2023ASR451TTS7Classification57Text-to-audio1Audio-to-audio1Audio-text-to-text0VAD1all audio518Apr 2023ASR351TTS11Classification60Text-to-audio13Audio-to-audio8Audio-text-to-text0VAD3all audio446May 2023ASR459TTS17Classification76Text-to-audio37Audio-to-audio9Audio-text-to-text0VAD0all audio598Jun 2023ASR412TTS13Classification111Text-to-audio66Audio-to-audio21Audio-text-to-text0VAD3all audio626Jul 2023ASR499TTS76Classification254Text-to-audio92Audio-to-audio24Audio-text-to-text0VAD2all audio947Aug 2023ASR1,144TTS102Classification213Text-to-audio73Audio-to-audio66Audio-text-to-text0VAD0all audio1,598Sep 2023ASR675TTS1,163Classification131Text-to-audio50Audio-to-audio98Audio-text-to-text0VAD3all audio2,120Oct 2023ASR908TTS22Classification74Text-to-audio54Audio-to-audio20Audio-text-to-text0VAD2all audio1,080Nov 2023ASR579TTS34Classification97Text-to-audio33Audio-to-audio13Audio-text-to-text0VAD1all audio757Dec 2023ASR655TTS50Classification96Text-to-audio40Audio-to-audio19Audio-text-to-text0VAD0all audio860Jan 2024ASR672TTS31Classification176Text-to-audio62Audio-to-audio17Audio-text-to-text0VAD1all audio959Feb 2024ASR622TTS39Classification127Text-to-audio53Audio-to-audio16Audio-text-to-text0VAD3all audio860Mar 2024ASR595TTS52Classification156Text-to-audio44Audio-to-audio28Audio-text-to-text0VAD1all audio876Apr 2024ASR746TTS47Classification155Text-to-audio124Audio-to-audio19Audio-text-to-text0VAD3all audio1,094May 2024ASR996TTS27Classification162Text-to-audio125Audio-to-audio17Audio-text-to-text0VAD0all audio1,327Jun 2024ASR964TTS26Classification136Text-to-audio147Audio-to-audio39Audio-text-to-text1VAD0all audio1,313Jul 2024ASR910TTS33Classification147Text-to-audio69Audio-to-audio41Audio-text-to-text3VAD4all audio1,207Aug 2024ASR343TTS28Classification54Text-to-audio36Audio-to-audio49Audio-text-to-text2VAD2all audio514Sep 2024ASR456TTS31Classification44Text-to-audio56Audio-to-audio28Audio-text-to-text4VAD4all audio623Oct 2024ASR519TTS51Classification50Text-to-audio137Audio-to-audio5Audio-text-to-text5VAD1all audio768Nov 2024ASR610TTS65Classification76Text-to-audio140Audio-to-audio24Audio-text-to-text11VAD3all audio929Dec 2024ASR581TTS101Classification76Text-to-audio92Audio-to-audio5Audio-text-to-text7VAD3all audio865Jan 2025ASR473TTS93Classification85Text-to-audio132Audio-to-audio35Audio-text-to-text5VAD2all audio825Feb 2025ASR590TTS78Classification93Text-to-audio106Audio-to-audio21Audio-text-to-text12VAD2all audio902Mar 2025ASR659TTS264Classification70Text-to-audio100Audio-to-audio14Audio-text-to-text8VAD2all audio1,117Apr 2025ASR580TTS178Classification86Text-to-audio96Audio-to-audio11Audio-text-to-text9VAD3all audio963May 2025ASR617TTS104Classification99Text-to-audio165Audio-to-audio11Audio-text-to-text17VAD2all audio1,015Jun 2025ASR571TTS88Classification70Text-to-audio91Audio-to-audio13Audio-text-to-text8VAD1all audio842Jul 2025ASR488TTS110Classification71Text-to-audio97Audio-to-audio8Audio-text-to-text18VAD4all audio796Aug 2025ASR483TTS108Classification70Text-to-audio46Audio-to-audio17Audio-text-to-text11VAD0all audio735Sep 2025ASR483TTS117Classification78Text-to-audio46Audio-to-audio21Audio-text-to-text8VAD31all audio784Oct 2025ASR334TTS95Classification129Text-to-audio74Audio-to-audio25Audio-text-to-text19VAD3all audio679Nov 2025ASR425TTS97Classification128Text-to-audio34Audio-to-audio10Audio-text-to-text5VAD3all audio702Dec 2025ASR617TTS180Classification78Text-to-audio48Audio-to-audio33Audio-text-to-text5VAD4all audio965Jan 2026ASR411TTS203Classification44Text-to-audio43Audio-to-audio37Audio-text-to-text12VAD8all audio758Feb 2026ASR674TTS265Classification83Text-to-audio78Audio-to-audio30Audio-text-to-text10VAD22all audio1,162Mar 2026ASR818TTS371Classification112Text-to-audio66Audio-to-audio72Audio-text-to-text8VAD13all audio1,460Apr 2026ASR1,227TTS479Classification143Text-to-audio75Audio-to-audio47Audio-text-to-text28VAD14all audio2,013May 2026ASR1,617TTS662Classification113Text-to-audio200Audio-to-audio95Audio-text-to-text21VAD15all audio2,723Jun 2026ASR1,216TTS499Classification159Text-to-audio98Audio-to-audio66Audio-text-to-text21VAD23all audio2,082Jul 2026ASR1,297TTS477Classification129Text-to-audio108Audio-to-audio107Audio-text-to-text19VAD19all audio2,156Aug 2026ASR1,156TTS602Classification182Text-to-audio130Audio-to-audio112Audio-text-to-text38VAD24all audio2,244Sep 2026ASR1,398TTS664Classification195Text-to-audio115Audio-to-audio78Audio-text-to-text71VAD52all audio2,573Oct 2026ASR1,517TTS922Classification326Text-to-audio120Audio-to-audio180Audio-text-to-text39VAD50all audio3,154

The current month is partial — 25 of 31 days.

Each field on its own scale

A shared axis would flatten six of these into nothing: ASR is 130× VAD. Each sparkline is scaled to its own field, so what you compare is shape, not size. Growth is the last 12 months against the 12 before.

Who ships what the field actually uses

A contribution grid weighted by engagement, not output: labs ranked by the likes their audio models have earned, months across, darker where a month's releases landed. Counting repos instead would rank format converters and checkpoint farms above every research lab — openai earns 14K likes from 8 models while facebook earns 5.9K from 110. Sibling repos are collapsed into one model, so facebook/mms-tts counts once rather than 1,140 times, one per language. The dot is the field each lab mostly works in; hover a row for its most-liked model.

lablikesmodels
2023202420252026
openai15K8
nvidia14K115
pyannote11K7
microsoft8.6K13
hexgrad7.5K3
facebook6.4K110
Qwen6.1K12
fishaudio5.1K8
coqui4.3K2
ResembleAI3.3K10
nari-labs3.3K4
stabilityai3.3K11
openbmb2.9K3
ACE-Step2.6K15
mistralai2.6K4
OpenMOSS-Team2.5K21
sesame2.5K1
Edge02.1K14
Supertone1.9K3
suno1.8K2
FunAudioLLM1.8K16
Zyphra1.7K5
2Noise1.7K1
neuphonic1.7K19
16,224 labs have shipped audio models and earned 207,098 likes between them; these are the 24 the community engaged with mostfewer likes more

The landmarks, all on one timeline

The most-liked release from each field, ringed in its colour. Hover for the model, click to open it on the Hub.

Classificationwav2vec2-large-robust-12-faudeering/wav2vec2-large-robust-12-ft-emotion-msp-dim2022-04-06170K · 175 likesClassificationast-finetuned-audioset-10-MIT/ast-finetuned-audioset-10-10-0.45932022-11-141.4M · 384 likesASRwhisper-large-v2openai/whisper-large-v22022-12-0571K · 1,825 likesVADsegmentation-3.0pyannote/segmentation-3.02023-09-225.2M · 2,366 likesTTSXTTS-v2coqui/XTTS-v22023-10-316.2M · 3,911 likesASRwhisper-large-v3openai/whisper-large-v32023-11-073.6M · 6,689 likesASRspeaker-diarization-3.1pyannote/speaker-diarization-3.12023-11-166.9M · 4,602 likesText-to-audiostable-audio-open-1.0stabilityai/stable-audio-open-1.02024-05-2421K · 1,817 likesText-to-audioChatTTS2Noise/ChatTTS2024-05-254.2K · 1,688 likesAudio-text-to-textQwen2-Audio-7BQwen/Qwen2-Audio-7B-Instruct2024-07-31178K · 574 likesASRwhisper-large-v3-turboopenai/whisper-large-v3-turbo2024-10-015.9M · 3,476 likesAudio-to-audiofish-agent-v0.1fishaudio/fish-agent-v0.1-3b2024-10-2971 · 277 likesTTSKokoro-82Mhexgrad/Kokoro-82M2024-12-2610.3M · 7,271 likesAudio-text-to-textStep-Audio-Chatstepfun-ai/Step-Audio-Chat2025-02-13412 · 461 likesASRPhi-4-multimodalmicrosoft/Phi-4-multimodal-instruct2025-02-24117K · 1,626 likesTTSCSM-1Bsesame/csm-1b2025-03-06128K · 2,483 likesASRspeaker-diarization-communpyannote/speaker-diarization-community-12025-04-155M · 3,345 likesASRparakeet-tdt-0.6b-v2nvidia/parakeet-tdt-0.6b-v22025-04-15184K · 1,582 likesTTSDia-1.6Bnari-labs/Dia-1.6B2025-04-2016K · 2,933 likesAudio-text-to-textVoxtral-Small-24B-2507mistralai/Voxtral-Small-24B-25072025-07-1517K · 553 likesTTSVibeVoicemicrosoft/VibeVoice-1.5B2025-08-25365K · 2,506 likesVADsmart-turn-v3pipecat-ai/smart-turn-v32025-09-110 · 214 likesAudio-to-audioLFM2-AudioLiquidAI/LFM2-Audio-1.5B2025-09-19373 · 365 likesAudio-to-audiopersonaplex-7b-v1nvidia/personaplex-7b-v12025-12-3198K · 2,897 likesAudio-to-audioLFM2.5-AudioLiquidAI/LFM2.5-Audio-1.5B2026-01-056.7K · 472 likesASRVibeVoice-ASRmicrosoft/VibeVoice-ASR2026-01-21669K · 1,320 likesASRnemotron-3.5-asr-streamingnvidia/nemotron-3.5-asr-streaming-0.6b2026-05-151.1M · 1,232 likesAudio-text-to-textMOSS-Transcribe-DiarizeOpenMOSS-Team/MOSS-Transcribe-Diarize2026-05-19183K · 498 likesText-to-audioMiniMax-Music3MiniMaxAI/MiniMax-Music32026-08-075.9K · 1,474 likesVADNemotron-3-Diarizationnvidia/Nemotron-3-Diarization2026-09-01118K · 805 likesText-to-audioYuE2m-a-p/YuE2-3B2026-09-0925K · 1,197 likes
2023202420252026
Table view — every figure on this page
Every figure on this page, as text.
TaskModelsShareLast 12moPrior 12moGrowthLandmarks
ASR38,30965.2%11,1906,6541.68×40
TTS8,92815.2%4,5941,3573.39×63
Classification5,3359.1%1,4959241.62×52
Text-to-audio3,6256.2%1,0691,2480.86×50
Audio-to-audio1,7523.0%7121853.85×52
Audio-text-to-text4250.7%2571192.16×28
VAD3480.6%200543.70×20