Open ASR · AMI-Cleaned WER ↗English short-form; normalization unversioned
7.19% WERbosonai/higgs-audio-v3-stt
bosonaiDiscovered Oct 17, 2026
bosonai/higgs-audio-v3-stt
2.68B parameters · 1 language · Encoder: Whisper · Decoder: Qwen3
Measured performance
All leaderboards ↗Benchmark results
Word error rate · lower is better
Open ASR · Earnings22-Cleaned-AA-chunked WER ↗English short-form; normalization unversioned
7.04% WERbosonai/higgs-audio-v3-stt
Open ASR · Gigaspeech-Cleaned WER ↗English short-form; normalization unversioned
7.45% WERbosonai/higgs-audio-v3-stt
Open ASR · LS Clean WER ↗English short-form; normalization unversioned
1.08% WERbosonai/higgs-audio-v3-stt
Open ASR · LS Other WER ↗English short-form; normalization unversioned
2.61% WERbosonai/higgs-audio-v3-stt
Open ASR · SPGISpeech WER ↗English short-form; normalization unversioned
2.07% WERbosonai/higgs-audio-v3-stt
Open ASR · URGENT2024 WER ↗English short-form; normalization unversioned
11.88% WERbosonai/higgs-audio-v3-stt
Open ASR · Voice Arena Monsoon WER ↗English short-form; normalization unversioned
3.97% WERbosonai/higgs-audio-v3-stt
Open ASR · Voxpopuli-AA-Cleaned WER ↗English short-form; normalization unversioned
3.73% WERbosonai/higgs-audio-v3-stt
Access & cost
0 quotesPrices across providers
No listed provider price
Model specifications
- Developer
- bosonai
- Modality
- Audio
- Context
- Not reported
- Maximum output
- Not reported
- Weights
- Not reported
- First discovered
- Oct 17, 2026
Exact identities
- open-asr
bosonai%2Fhiggs-audio-v3-stt
Sources
Updated Oct 17, 2026
More from bosonai
View all ↗bosonaiListed Jan 16
HiggsAudioV2.5
HiggsAudioV2.5 is a high-quality neural text-to-speech (TTS) model designed for natural-sounding voice generation across a wide range of use cases. It focuses on clarity, stable prosody, and consistent…
No reported score
No reported score