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0d32e02c79d7
diff --git a/funasr/models/lcbnet/model.py b/funasr/models/lcbnet/model.py
index d1ebc5ce5..1acce785f 100644
--- a/funasr/models/lcbnet/model.py
+++ b/funasr/models/lcbnet/model.py
@@ -425,14 +425,17 @@ class LCBNet(nn.Module):
# extract fbank feats
time1 = time.perf_counter()
pdb.set_trace()
- audio_sample_list = load_audio_text_image_video(data_in, fs=frontend.fs, audio_fs=kwargs.get("fs", 16000),
+ sample_list = load_audio_text_image_video(data_in, fs=frontend.fs, audio_fs=kwargs.get("fs", 16000),
data_type=kwargs.get("data_type", "sound"),
tokenizer=tokenizer)
time2 = time.perf_counter()
meta_data["load_data"] = f"{time2 - time1:0.3f}"
+ audio_sample_list = sample_list[0]
+ ocr_sample_list = sample_list[1]
pdb.set_trace()
speech, speech_lengths = extract_fbank(audio_sample_list, data_type=kwargs.get("data_type", "sound"),
frontend=frontend)
+ pdb.set_trace()
time3 = time.perf_counter()
meta_data["extract_feat"] = f"{time3 - time2:0.3f}"
meta_data["batch_data_time"] = speech_lengths.sum().item() * frontend.frame_shift * frontend.lfr_n / 1000