From 6e47d42ea00e6d10746b59a86d6455465464ed83 Mon Sep 17 00:00:00 2001
From: 语帆 <yf352572@alibaba-inc.com>
Date: 星期三, 28 二月 2024 15:55:23 +0800
Subject: [PATCH] test
---
funasr/models/lcbnet/model.py | 9 ++++++---
1 files changed, 6 insertions(+), 3 deletions(-)
diff --git a/funasr/models/lcbnet/model.py b/funasr/models/lcbnet/model.py
index f8bbf7a..3b8f3c9 100644
--- a/funasr/models/lcbnet/model.py
+++ b/funasr/models/lcbnet/model.py
@@ -296,7 +296,6 @@
if intermediate_outs is not None:
return (encoder_out, intermediate_outs), encoder_out_lens
- pdb.set_trace()
return encoder_out, encoder_out_lens
def _calc_att_loss(
@@ -423,6 +422,7 @@
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)
@@ -439,12 +439,15 @@
speech = speech.to(device=kwargs["device"])
speech_lengths = speech_lengths.to(device=kwargs["device"])
- pdb.set_trace()
# Encoder
encoder_out, encoder_out_lens = self.encode(speech, speech_lengths)
if isinstance(encoder_out, tuple):
encoder_out = encoder_out[0]
-
+ pdb.set_trace()
+ ocr = ocr_sample_list[0]
+ ocr_lengths = ocr.new_full([1], dtype=torch.long, fill_value=ocr.size(1))
+ ocr, ocr_lens, _ = self.text_encoder(ocr, ocr_lengths)
+ pdb.set_trace()
# c. Passed the encoder result and the beam search
nbest_hyps = self.beam_search(
x=encoder_out[0], maxlenratio=kwargs.get("maxlenratio", 0.0), minlenratio=kwargs.get("minlenratio", 0.0)
--
Gitblit v1.9.1