Skip to content

huggingface_models.md

Latest commit

 

History

History
155 lines (88 loc) · 16.3 KB

huggingface_models.md

File metadata and controls

155 lines (88 loc) · 16.3 KB
Raw

Pretrained Models on Huggingface

Model License

The FunASR toolkit uses the MIT software license. Model weights have separate terms: check the exact checkpoint's model card, license file and revision before use or redistribution. The FunASR model agreement applies only to models whose published terms adopt it, not every entry in this catalogue. Third-party models, including OpenMOSS MOSS-Transcribe-Diarize, retain their original authorship and model-specific licenses.

For model selection and deployment boundaries, start with the Model Zoo guide and MOSS integration.

Model Zoo

Here we provided several pretrained models on different datasets. The details of models and datasets can be found on ModelScope.

Speech Recognition Models

Fun-ASR-Nano: choose the checkpoint for your interface

Checkpoint Interface Scope
Fun-ASR-Nano-2512-hf Native Transformers 5.17.0, AutoProcessor + AutoModelForSpeechSeq2Seq Chinese, English, Japanese transcription; no remote Python code, native word timestamps or speaker identities
Fun-ASR-Nano-2512 FunASR toolkit AutoModel Original toolkit artifact; not interchangeable with the native export
Fun-ASR-Nano-2512-vllm Native vLLM Separate serving conversion and runtime
Fun-ASR-MLT-Nano-2512 FunASR toolkit Separate 31-language checkpoint, not the zh/en/ja native export

Start with the native Transformers guide, online Space, or batch examples and notebook. For a service, follow the deployment matrix.

Paraformer Models

Model Name Language Training Data Vocab Size Parameter Offline/Online Notes
Paraformer-large CN & EN Alibaba Speech Data (60000hours) 8404 220M Offline Duration of input wav <= 20s

UniASR Models

Conformer Models

RNN-T Models

Multi-talker Speech Recognition Models

MFCCA Models

Voice Activity Detection Models

Model Name Training Data Parameters Sampling Rate Notes
FSMN-VAD Alibaba Speech Data (5000hours) 0.4M 16000

Punctuation Restoration Models

Model Name Training Data Parameters Vocab Size Offline/Online Notes
CT-Transformer Alibaba Text Data 70M 272727 Offline offline punctuation model

Language Models

Speaker Verification Models

Speaker diarization Models

Third-party Unified Transcription and Diarization

Model Name Publisher Parameters Notes
MOSS-Transcribe-Diarize OpenMOSS See official model card Third-party Apache-2.0 model. One offline request returns transcription, timestamps, and speaker labels; the FunASR integration does not require an external VAD service. See the deployment guide.

Timestamp Prediction Models