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
fix: support loading .pcm (16k 1c 16bit) audio files in load_utils.py (#2667) (#2668)
weilikaieb1574b813e2
diff --git a/funasr/utils/load_utils.py b/funasr/utils/load_utils.pyindex c38f1d937..d208f7dc0 100644--- a/funasr/utils/load_utils.py+++ b/funasr/utils/load_utils.py@@ -241,10 +241,19 @@ def _load_audio_ffmpeg(file: str, sr: int = 16000):# This launches a subprocess to decode audio while down-mixing# and resampling as necessary. Requires the ffmpeg CLI in PATH.# fmt: off+ pcm_params = []+ if file.lower().endswith('.pcm'):+ pcm_params = [+ "-f", "s16le",+ "-ar", str(sr),+ "-ac", "1"+ ]+cmd = ["ffmpeg","-nostdin","-threads", "0",+ *pcm_params, # PCM files need input format specified before -i since PCM is raw data without headers"-i", file,"-f", "s16le","-ac", "1",