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 bug
游雁82530ddf974a
diff --git a/examples/industrial_data_pretraining/llm_asr/demo_speech2text.sh b/examples/industrial_data_pretraining/llm_asr/demo_speech2text.shindex c4e7afe88..d7bd1cb22 100644--- a/examples/industrial_data_pretraining/llm_asr/demo_speech2text.sh+++ b/examples/industrial_data_pretraining/llm_asr/demo_speech2text.sh@@ -1,6 +1,4 @@--ckpt_dir="/nfs/beinian.lzr/workspace/GPT-4o/Exp/exp6/5m-8gpu/exp6_speech2text_linear_ddp_0609"ckpt_id="model.pt.ep0.90000"jsonl_dir="/nfs/beinian.lzr/workspace/GPT-4o/Data/Speech2Text/TestData"@@ -9,6 +7,9 @@ mkdir -p ${out_dir}device="cuda:0"+ckpt_id=$1+device=$2+for data_set in "librispeech_test_clean_speech2text.jsonl" "librispeech_test_other_speech2text.jsonl"; dojsonl=${jsonl_dir}/${data_set}output_dir=${out_dir}/${data_set}