From 2de29621c5b75fac8324ab9e04f0ce59c09a86b6 Mon Sep 17 00:00:00 2001
From: zhifu gao <zhifu.gzf@alibaba-inc.com>
Date: 星期三, 18 一月 2023 13:52:01 +0800
Subject: [PATCH] Merge pull request #37 from alibaba-damo-academy/dev_lyb
---
funasr/bin/asr_inference_paraformer_vad_punc.py | 24 +++++++++++++++---------
1 files changed, 15 insertions(+), 9 deletions(-)
diff --git a/funasr/bin/asr_inference_paraformer_vad_punc.py b/funasr/bin/asr_inference_paraformer_vad_punc.py
index 7752ea9..85838aa 100644
--- a/funasr/bin/asr_inference_paraformer_vad_punc.py
+++ b/funasr/bin/asr_inference_paraformer_vad_punc.py
@@ -100,10 +100,13 @@
# logging.info("asr_train_args: {}".format(asr_train_args))
asr_model.to(dtype=getattr(torch, dtype)).eval()
- ctc = CTCPrefixScorer(ctc=asr_model.ctc, eos=asr_model.eos)
+ if asr_model.ctc != None:
+ ctc = CTCPrefixScorer(ctc=asr_model.ctc, eos=asr_model.eos)
+ scorers.update(
+ ctc=ctc
+ )
token_list = asr_model.token_list
scorers.update(
- ctc=ctc,
length_bonus=LengthBonus(len(token_list)),
)
@@ -562,6 +565,7 @@
length_total = 0.0
finish_count = 0
file_count = 1
+ lfr_factor = 6
# 7 .Start for-loop
asr_result_list = []
output_path = output_dir_v2 if output_dir_v2 is not None else output_dir
@@ -597,7 +601,7 @@
results = speech2text(**batch)
if len(results) < 1:
hyp = Hypothesis(score=0.0, scores={}, states={}, yseq=[])
- results = [[" ", ["<space>"], [2], 10, 6]] * nbest
+ results = [[" ", ["<space>"], [2], 0, 1, 6]] * nbest
time_end = time.time()
forward_time = time_end - time_beg
lfr_factor = results[0][-1]
@@ -615,7 +619,8 @@
key = keys[0]
result = result_segments[0]
- text, token, token_int, time_stamp = result
+ text, token, token_int = result[0], result[1], result[2]
+ time_stamp = None if len(result) < 4 else result[3]
# Create a directory: outdir/{n}best_recog
if writer is not None:
@@ -634,11 +639,12 @@
text_postprocessed_punc_time_stamp = "predictions: {} time_stamp: {}".format(
text_postprocessed_punc, time_stamp_postprocessed)
else:
- text_postprocessed = postprocessed_result
- time_stamp_postprocessed = None
- word_lists = None
- text_postprocessed_punc_time_stamp = None
- punc_id_list = None
+ text_postprocessed = ""
+ time_stamp_postprocessed = ""
+ word_lists = ""
+ text_postprocessed_punc_time_stamp = ""
+ punc_id_list = ""
+ text_postprocessed_punc = ""
item = {'key': key, 'value': text_postprocessed_punc_time_stamp, 'text': text_postprocessed,
'time_stamp': time_stamp_postprocessed, 'punc': punc_id_list, 'token': token}
--
Gitblit v1.9.1