FunASR Series

Paraformer Chinese-English Models

Model NameTypeLanguagesPunctuationTimestampsDownload Link
paraformer-large-zh-en-onnx-offlineNon-streamingChinese (zh), English (en)NoNohuggingface, modelscope
paraformer-large-zh-en-timestamp-onnx-offlineNon-streamingChinese, EnglishNoYesmodelscope
paraformer-large-en-onnx-offlineNon-streamingEnglishNoNomodelscope
paraformer-large-zh-en-onnx-onlineStreamingChinese, EnglishNoNomodelscope

Paraformer Cantonese/Chinese/English Multilingual Models

Model NameTypeLanguagesPunctuationTimestampsDownload Link
paraformer-large-zh-yue-en-timestamp-onnx-offline-dengcunqin-20240805Non-streamingChinese, Cantonese, EnglishNoYesmodelscope
paraformer-large-zh-yue-en-onnx-offline-dengcunqin-20240805Non-streamingChinese, Cantonese, EnglishNoNomodelscope
paraformer-large-zh-yue-en-onnx-online-dengcunqin-20240208StreamingChinese, Cantonese, EnglishNoNomodelscope

SeACo-Paraformer Hotword Customization Model

SeACoParaformer is a next-generation non-autoregressive speech recognition model with hotword customization, proposed by Alibaba Speech Lab. Compared to the previous CLAS-based hotword customization solution, SeACoParaformer decouples the hotword module from the ASR model and performs hotword boosting via posterior probability fusion, making the boosting process visible and controllable, while significantly improving hotword recall.

Model NameTypeLanguagesPunctuationTimestampsDownload Link
paraformer-seaco-large-zh-timestamp-onnx-offlineNon-streamingChinese, supports hotword customizationNoYesmodelscope

SenseVoice Models

Model NameTypeLanguagesPunctuationTimestampsDownload Link
sensevoice-small-onnxNon-streamingChinese, Cantonese, English, Japanese, KoreanYesNomodelscope
sensevoice-small-int8-onnxNon-streamingChinese, Cantonese, English, Japanese, KoreanYesNomodelscope
sensevoice-small-wenetspeech-yue-onnxNon-streamingCantonese, Chinese, English, Japanese, KoreanYesNomodelscope
sensevoice-small-wenetspeech-yue-int8-onnxNon-streamingCantonese, Chinese, English, Japanese, KoreanYesNomodelscope
sensevoice-small-split-embed-onnxNon-streamingChinese, Cantonese, English, Japanese, KoreanYesNomodelscope