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d5c98e6f65cf
diff --git a/examples/aishell/branchformer/run.sh b/examples/aishell/branchformer/run.shindex 4e51288f2..ecdfd24c9 100755--- a/examples/aishell/branchformer/run.sh+++ b/examples/aishell/branchformer/run.sh@@ -78,7 +78,7 @@ if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then--config-name "${config}" \++train_data_set_list="${feats_dir}/data/${train_set}/audio_datasets.jsonl" \++cmvn_file="${feats_dir}/data/${train_set}/cmvn.json" \- ++dataset_conf.num_workers=$nj+fitoken_list=${feats_dir}/data/${lang}_token_list/$token_type/tokens.txtdiff --git a/examples/aishell/conformer/run.sh b/examples/aishell/conformer/run.shindex f5d993a28..19aa2db6a 100755--- a/examples/aishell/conformer/run.sh+++ b/examples/aishell/conformer/run.sh@@ -77,7 +77,7 @@ if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then--config-name "${config}" \++train_data_set_list="${feats_dir}/data/${train_set}/audio_datasets.jsonl" \++cmvn_file="${feats_dir}/data/${train_set}/cmvn.json" \- ++dataset_conf.num_workers=$nj+fitoken_list=${feats_dir}/data/${lang}_token_list/$token_type/tokens.txtdiff --git a/examples/aishell/e_branchformer/run.sh b/examples/aishell/e_branchformer/run.shindex 5f8fcfbff..e5a8569ff 100755--- a/examples/aishell/e_branchformer/run.sh+++ b/examples/aishell/e_branchformer/run.sh@@ -78,7 +78,7 @@ if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then--config-name "${config}" \++train_data_set_list="${feats_dir}/data/${train_set}/audio_datasets.jsonl" \++cmvn_file="${feats_dir}/data/${train_set}/cmvn.json" \- ++dataset_conf.num_workers=$nj+fitoken_list=${feats_dir}/data/${lang}_token_list/$token_type/tokens.txtdiff --git a/examples/aishell/paraformer/run.sh b/examples/aishell/paraformer/run.shindex 6200a4af5..3c7feb2a1 100755--- a/examples/aishell/paraformer/run.sh+++ b/examples/aishell/paraformer/run.sh@@ -77,7 +77,6 @@ if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then--config-name "${config}" \++train_data_set_list="${feats_dir}/data/${train_set}/audio_datasets.jsonl" \++cmvn_file="${feats_dir}/data/${train_set}/cmvn.json" \- ++dataset_conf.num_workers=$njfitoken_list=${feats_dir}/data/${lang}_token_list/$token_type/tokens.txtdiff --git a/examples/aishell/transformer/run.sh b/examples/aishell/transformer/run.shindex 575abf871..e9c3d6414 100755--- a/examples/aishell/transformer/run.sh+++ b/examples/aishell/transformer/run.sh@@ -78,7 +78,7 @@ if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then--config-name "${config}" \++train_data_set_list="${feats_dir}/data/${train_set}/audio_datasets.jsonl" \++cmvn_file="${feats_dir}/data/${train_set}/cmvn.json" \- ++dataset_conf.num_workers=$nj+fitoken_list=${feats_dir}/data/${lang}_token_list/$token_type/tokens.txt