Skip to content

Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.

test

语帆d12d18886cf5
diff --git a/examples/industrial_data_pretraining/lcbnet/demo_pdb.sh b/examples/industrial_data_pretraining/lcbnet/demo_pdb.sh
new file mode 100644
index 000000000..e435905bf
--- /dev/null
+++ b/examples/industrial_data_pretraining/lcbnet/demo_pdb.sh
@@ -0,0 +1,13 @@
+file_dir="/nfs/yufan.yf/workspace/github/FunASR/examples/industrial_data_pretraining/lcbnet/exp/speech_lcbnet_contextual_asr-en-16k-bpe-vocab5002-pytorch"
+
+#CUDA_VISIBLE_DEVICES="" \
+python -m funasr.bin.inference \
+--config-path=${file_dir} \
+--config-name="config.yaml" \
+++init_param=${file_dir}/model.pb \
+++tokenizer_conf.token_list=${file_dir}/tokens.txt \
+++input=[${file_dir}/dev/wav.scp,${file_dir}/dev/ocr.txt] \
++data_type='["kaldi_ark", "text"]' \
+++tokenizer_conf.bpemodel=${file_dir}/bpe.model \
+++output_dir="./outputs/debug" \
+++device="cpu" \