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
time_speech_total_per_sample=0, bugfix
游雁ca1f8ee4230b
diff --git a/funasr/auto/auto_model.py b/funasr/auto/auto_model.pyindex 5bfc08040..e774a8ffc 100644--- a/funasr/auto/auto_model.py+++ b/funasr/auto/auto_model.py@@ -311,7 +311,7 @@ class AutoModel:batch_size_ms_cum = 0beg_idx = 0beg_asr_total = time.time()- time_speech_total_per_sample = speech_lengths/16000+ time_speech_total_per_sample = speech_lengths/16000 + 1e-6time_speech_total_all_samples += time_speech_total_per_samplepbar_sample = tqdm(colour="blue", total=n + 1, dynamic_ncols=True)