Hugging Face
Models
Datasets
Spaces
Buckets
new
Docs
Enterprise
Pricing
Log In
Sign Up
bosonai
/
higgs-audio-v3-stt
like
10
Follow
BosonAI
393
Automatic Speech Recognition
Safetensors
English
higgs_audio_3
hf-asr-leaderboard
whisper
qwen
custom_code
Eval Results (legacy)
License:
apache-2.0
Model card
Files
Files and versions
xet
Community
2
main
higgs-audio-v3-stt
5.37 GB
Ctrl+K
Ctrl+K
1 contributor
History:
26 commits
erik-at-boson
Add leaderboard reproduction helpers
db49668
verified
12 days ago
open_asr_leaderboard
Add leaderboard reproduction helpers
12 days ago
.gitattributes
Safe
1.57 kB
Upload folder using huggingface_hub
about 1 month ago
README.md
Safe
7.43 kB
Update model weights and eval results (avg WER 5.67% → 5.36%)
21 days ago
added_tokens.json
Safe
1.19 kB
Upload folder using huggingface_hub
about 1 month ago
attention.py
Safe
8.71 kB
Fix: use importlib for deepspeed to avoid static import check
about 1 month ago
common.py
Safe
1.09 kB
fix: dual-mode imports + complete ChatMLDatasetSample + transcribe output parsing
17 days ago
config.json
Safe
5.91 kB
fix: add auto_map + make xcodec imports lazy for trust_remote_code support
18 days ago
configuration_higgs_audio.py
Safe
9.88 kB
Fix: handle higgs_audio_encoder config without global registry
about 1 month ago
cuda_graph_runner.py
Safe
5.14 kB
Add model code for trust_remote_code support
about 1 month ago
custom_modules.py
Safe
6.19 kB
Add model code for trust_remote_code support
about 1 month ago
generation_config.json
Safe
147 Bytes
Upload folder using huggingface_hub
about 1 month ago
higgs_audio_collator.py
Safe
36.6 kB
fix: dual-mode imports + complete ChatMLDatasetSample + transcribe output parsing
17 days ago
merges.txt
Safe
1.67 MB
Upload folder using huggingface_hub
about 1 month ago
model-00001-of-00002.safetensors
Safe
4.73 GB
xet
Update model weights and eval results (avg WER 5.67% → 5.36%)
21 days ago
model-00002-of-00002.safetensors
Safe
622 MB
xet
Upload folder using huggingface_hub
about 1 month ago
model.safetensors.index.json
Safe
65.7 kB
Upload folder using huggingface_hub
about 1 month ago
modeling_higgs_audio.py
Safe
110 kB
feat: batched generation for ASR (remove bs=1 assertion when no audio_out tokens)
16 days ago
modeling_higgs_audio_xcodec.py
Safe
20.2 kB
fix: dual-mode imports + complete ChatMLDatasetSample + transcribe output parsing
17 days ago
special_tokens_map.json
Safe
613 Bytes
Upload folder using huggingface_hub
about 1 month ago
tokenizer.json
Safe
11.4 MB
xet
Upload folder using huggingface_hub
about 1 month ago
tokenizer_config.json
Safe
12.7 kB
Upload folder using huggingface_hub
about 1 month ago
transcribe.py
Safe
14.5 kB
Embed internal leaderboard-parity eval logic in transcribe.py
12 days ago
utils.py
Safe
36.5 kB
Fix: use importlib for deepspeed to avoid static import check
about 1 month ago
vocab.json
Safe
2.78 MB
Upload folder using huggingface_hub
about 1 month ago