speech_recognition/FunASR
1
0
Fork 0
mirror of https://github.com/modelscope/FunASR synced 2025-09-15 14:48:36 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

test

Browse Source
This commit is contained in:
语帆 2024-02-28 15:05:41 +08:00
parent 2bffe1d539
commit 19103386dc
1 changed files with 2 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
funasr/models/lcbnet/model.py
View File

@ -426,7 +426,6 @@ class LCBNet(nn.Module):
else: else:
# extract fbank feats # extract fbank feats
time1 = time.perf_counter() time1 = time.perf_counter()
pdb.set_trace()
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"), data_type=kwargs.get("data_type", "sound"),
tokenizer=tokenizer) tokenizer=tokenizer)
@ -434,13 +433,12 @@ class LCBNet(nn.Module):
meta_data["load_data"] = f"{time2 - time1:0.3f}" meta_data["load_data"] = f"{time2 - time1:0.3f}"
audio_sample_list = sample_list[0] audio_sample_list = sample_list[0]
ocr_sample_list = sample_list[1] ocr_sample_list = sample_list[1]
pdb.set_trace()
speech, speech_lengths = extract_fbank(audio_sample_list, data_type=kwargs.get("data_type", "sound"), speech, speech_lengths = extract_fbank(audio_sample_list, data_type=kwargs.get("data_type", "sound"),
frontend=frontend) frontend=frontend)
pdb.set_trace()
time3 = time.perf_counter() time3 = time.perf_counter()
meta_data["extract_feat"] = f"{time3 - time2:0.3f}" 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 frame_shift = 10
meta_data["batch_data_time"] = speech_lengths.sum().item() * frame_shift / 1000
speech = speech.to(device=kwargs["device"]) speech = speech.to(device=kwargs["device"])
speech_lengths = speech_lengths.to(device=kwargs["device"]) speech_lengths = speech_lengths.to(device=kwargs["device"])