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
update
游雁d929c8e0f7bf
diff --git a/funasr/bin/compute_audio_cmvn.py b/funasr/bin/compute_audio_cmvn.pyindex 6282e707c..cd64329c3 100644--- a/funasr/bin/compute_audio_cmvn.py+++ b/funasr/bin/compute_audio_cmvn.py@@ -59,6 +59,7 @@ def main(**kwargs):dataset_conf = kwargs.get("dataset_conf")dataset_conf["batch_type"] = "example"dataset_conf["batch_size"] = 1+ dataset_conf["num_workers"] = os.cpu_count() or 32batch_sampler_train = batch_sampler_class(dataset_train, is_training=False, **dataset_conf)dataloader_train = torch.utils.data.DataLoader(dataset_train, collate_fn=dataset_train.collator, **batch_sampler_train)