Speech and audio trends · updated Oct 2, 2026

Which open-weight speech and audio models get downloaded most

A daily read of Hugging Face's top 5,000 models, focused on speech and audio models such as speech recognition and text to speech: the most downloaded, new uploads in the top 5,000, what they are used for, and who makes them.

Most downloaded

The most-downloaded open-weight speech and audio models

Ranked by downloads over the last 30 days. Fine-tuned versions count; most format conversions and re-uploads are left out. Counts include automated fetches, which lift models that tools and test suites pull by default. How models are placed is explained under About this data.

#	Model	Maker	Task	Licence	Downloads	Valkyrie
1	wav2vec2-large-xlsr-53-japanese	jonatasgrosman	Speech recognition	Apache-2.0	16.3M	Fine-tune one like it →Fine-tune one →
2	Kokoro-82M	hexgrad	Text to speech	Apache-2.0	11.5M	Fine-tune one like it →Fine-tune one →
3	clap-htsat-fused	laion	Audio classification	Apache-2.0	7.3M	Fine-tune one like it →Fine-tune one →
4	XTTS-v2	coqui	Text to speech	Custom licence	6.7M
5	whisper-large-v3-turbo	openai	Speech recognition	MIT	6.4M	Fine-tune one like it →Fine-tune one →
6	wav2vec2-large-xlsr-53-portuguese	jonatasgrosman	Speech recognition	Apache-2.0	5.7M	Fine-tune one like it →Fine-tune one →
7	speaker-diarization-community-1	pyannote	Speaker diarization	CC BY 4.0	5.5M	Fine-tune one like it →Fine-tune one →
8	segmentation-3.0	pyannote	Voice activity detection	MIT	5.4M	Fine-tune one like it →Fine-tune one →
9	wav2vec2-large-xlsr-53-russian	jonatasgrosman	Speech recognition	Apache-2.0	4.3M	Fine-tune one like it →Fine-tune one →
10	whisper-large-v3	openai	Speech recognition	Apache-2.0	4.2M	Fine-tune one like it →Fine-tune one →
11	whisper-small	openai	Speech recognition	Apache-2.0	3.2M	Fine-tune one like it →Fine-tune one →
12	mms-300m-1130-forced-aligner	MahmoudAshraf	Speech recognition	CC BY-NC 4.0	2.8M
13	Qwen3-TTS-12Hz-1.7B-CustomVoice	Qwen	Text to speech	Apache-2.0	2.4M	Fine-tune one like it →Fine-tune one →
14	wav2vec2-large-robust-24-ft-age-gender	audeering	Audio classification	CC BY-NC-SA 4.0	2.4M
15	wav2vec2-large-xlsr-53-polish	jonatasgrosman	Speech recognition	Apache-2.0	2.0M	Fine-tune one like it →Fine-tune one →
16	musicgen-medium	facebook	Text to audio	CC BY-NC 4.0	1.9M
17	Voxtral-Mini-4B-Realtime-2602	mistralai	Speech recognition	Apache-2.0	1.9M	Fine-tune one like it →Fine-tune one →
18	whisper-base	openai	Speech recognition	Apache-2.0	1.8M	Fine-tune one like it →Fine-tune one →
19	chatterbox	ResembleAI	Text to speech	MIT	1.7M	Fine-tune one like it →Fine-tune one →
20	wav2vec2-base-960h	facebook	Speech recognition	Apache-2.0	1.6M	Fine-tune one like it →Fine-tune one →
21	Qwen3-ASR-1.7B	Qwen	Speech recognition	Apache-2.0	1.6M	Fine-tune one like it →Fine-tune one →
22	OmniVoice	k2-fsa	Text to speech	-	1.4M
23	whisper-tiny	openai	Speech recognition	Apache-2.0	1.4M	Fine-tune one like it →Fine-tune one →
24	wav2vec2-large-xlsr-53-arabic	jonatasgrosman	Speech recognition	Apache-2.0	1.4M	Fine-tune one like it →Fine-tune one →
25	wav2vec2-large-xlsr-53-chinese-zh-cn	jonatasgrosman	Speech recognition	Apache-2.0	1.4M	Fine-tune one like it →Fine-tune one →

New releases

New speech and audio models in the top 5,000

Uploaded to Hugging Face in the last six months with at least 25 likes or a trending score of 10, ranked by downloads over the last 30 days. Most format conversions and re-uploads are left out.

nemotron-3.5-asr-streaming-0.6bnvidiaSpeech recognition · uploaded May 15, 20261.2M downloads in 30 days

VieNeu-TTS-v3-Turbopnnbao-umpText to speech · uploaded Jun 5, 2026761K downloads in 30 daysFine-tune one like it →Fine-tune one →

sanoTTSampixaText to speech · uploaded Jul 13, 2026411K downloads in 30 days

MOSS-TTS-v1.5OpenMOSS-TeamText to speech · uploaded May 25, 2026130K downloads in 30 daysFine-tune one like it →Fine-tune one →

granite-speech-4.1-2bibm-graniteSpeech recognition · uploaded Apr 16, 2026122K downloads in 30 daysFine-tune one like it →Fine-tune one →

granite-speech-4.1-2b-plusibm-graniteSpeech recognition · uploaded Apr 16, 2026103K downloads in 30 daysFine-tune one like it →Fine-tune one →

What they're used for

Downloads by task

Share of downloads of speech and audio models in Hugging Face's top 5,000 on Oct 2, 2026, leaving out most format conversions and re-uploads.

Speech recognition58.5%

Text to speech20.8%

Audio classification9.3%

Voice activity detection4.1%

Speaker diarization3.7%

Audio to audio2.1%

Text to audio1.5%

Who's winning

Share of downloads by maker

Each repo counts under the account that uploaded it, leaving out most format conversions and re-uploads. Change is in percentage points since Jul 4, 2026.

jonatasgrosman22.6% (+5.7 pts)

openai11.1% (0.0 pts)

pyannote7.3% (+1.4 pts)

hexgrad7.1% (+0.8 pts)

Qwen4.8% (+1.5 pts)

laion4.6% (−1.4 pts)

facebook4.5% (+0.9 pts)

coqui4.2% (0.0 pts)

Other33.9% (−9.0 pts)

Licences of the most-downloaded speech and audio models

Apache-2.0 61.8%MIT 14.1%Llama licence 0.1%Other or none 24.1%

As declared on each repo, leaving out most format conversions and re-uploads.