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
test
语帆b13b4c1bf5db
diff --git a/funasr/frontends/default.py b/funasr/frontends/default.pyindex ab5b73166..66d42f71c 100644--- a/funasr/frontends/default.py+++ b/funasr/frontends/default.py@@ -26,7 +26,7 @@ class DefaultFrontend(nn.Module):def __init__(self,- fs: Union[int, str] = 16000,+ fs: int = 16000,n_fft: int = 512,win_length: int = None,hop_length: int = 128,@@ -44,8 +44,6 @@ class DefaultFrontend(nn.Module):**kwargs,):super().__init__()- if isinstance(fs, str):- fs = humanfriendly.parse_size(fs)# Deepcopy (In general, dict shouldn't be used as default arg)frontend_conf = copy.deepcopy(frontend_conf)@@ -147,7 +145,7 @@ class MultiChannelFrontend(nn.Module):def __init__(self,- fs: Union[int, str] = 16000,+ fs: int = 16000,n_fft: int = 512,win_length: int = None,hop_length: int = None,@@ -170,9 +168,6 @@ class MultiChannelFrontend(nn.Module):mc: bool = True):super().__init__()- if isinstance(fs, str):- fs = humanfriendly.parse_size(fs)-# Deepcopy (In general, dict shouldn't be used as default arg)frontend_conf = copy.deepcopy(frontend_conf)if win_length is None and hop_length is None: