From 52fee96d71ba96fd09ad453dbae1926a1d601a56 Mon Sep 17 00:00:00 2001
From: 语帆 <yf352572@alibaba-inc.com>
Date: 星期三, 28 二月 2024 15:31:14 +0800
Subject: [PATCH] test
---
funasr/models/lcbnet/model.py | 9 ++-------
1 files changed, 2 insertions(+), 7 deletions(-)
diff --git a/funasr/models/lcbnet/model.py b/funasr/models/lcbnet/model.py
index f45e71d..f8bbf7a 100644
--- a/funasr/models/lcbnet/model.py
+++ b/funasr/models/lcbnet/model.py
@@ -274,15 +274,12 @@
ind: int
"""
with autocast(False):
- pdb.set_trace()
# Data augmentation
if self.specaug is not None and self.training:
speech, speech_lengths = self.specaug(speech, speech_lengths)
- pdb.set_trace()
# Normalization for feature: e.g. Global-CMVN, Utterance-CMVN
if self.normalize is not None:
speech, speech_lengths = self.normalize(speech, speech_lengths)
- pdb.set_trace()
# Forward encoder
# feats: (Batch, Length, Dim)
# -> encoder_out: (Batch, Length2, Dim2)
@@ -426,7 +423,6 @@
else:
# extract fbank feats
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),
data_type=kwargs.get("data_type", "sound"),
tokenizer=tokenizer)
@@ -434,13 +430,12 @@
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
+ frame_shift = 10
+ meta_data["batch_data_time"] = speech_lengths.sum().item() * frame_shift / 1000
speech = speech.to(device=kwargs["device"])
speech_lengths = speech_lengths.to(device=kwargs["device"])
--
Gitblit v1.9.1