Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.
Python asraudiochineseemotion-recognitionfunasrmcp-servermultilingual-asropenai-compatible-apiparaformerpunctuationpytorchreal-time-asrspeaker-diarizationspeech-recognitionspeech-to-textstreaming-asrtranscriptionvllmvoice-activity-detectionwhisper-alternative
ffmpeg
游雁ee4bf7fba4c3
diff --git a/funasr/utils/load_utils.py b/funasr/utils/load_utils.pyindex 484940874..5dc9bdea6 100644--- a/funasr/utils/load_utils.py+++ b/funasr/utils/load_utils.py@@ -30,7 +30,10 @@ use_ffmpeg=Falseif is_ffmpeg_installed():use_ffmpeg = Trueelse:- print("Notice: ffmpeg is not installed. torchaudio is used to load audio")+ print("Notice: ffmpeg is not installed. torchaudio is used to load audio\n"+ "If you want use ffmpeg backend to load audio, please install it by:"+ "\n\tsudo apt install ffmpeg # ubuntu"+ "\n\t# brew install ffmpeg # mac")def load_audio_text_image_video(data_or_path_or_list, fs: int = 16000, audio_fs: int = 16000, data_type="sound", tokenizer=None, **kwargs):if isinstance(data_or_path_or_list, (list, tuple)):