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

update paraformer online text postprocess

Browse Source
This commit is contained in:
haoneng.lhn 2023-05-19 13:19:55 +08:00
parent 025df72c10
commit dcc4f728cd
3 changed files with 50 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
egs_modelscope/asr/paraformer/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-online/demo.py
View File

@ -34,6 +34,6 @@ for sample_offset in range(0, speech_length, min(stride_size, speech_length - sa
rec_result = inference_pipeline(audio_in=speech[sample_offset: sample_offset + stride_size],
param_dict=param_dict)
if len(rec_result) != 0:
final_result += rec_result['text'] + " "
final_result += rec_result['text']
print(rec_result)
print(final_result)

39
egs_modelscope/asr/paraformer/speech_paraformer_asr_nat-zh-cn-16k-common-vocab8404-online/demo.py Normal file
View File

@ -0,0 +1,39 @@
import os
import logging
import torch
import soundfile
from modelscope.pipelines import pipeline
from modelscope.utils.constant import Tasks
from modelscope.utils.logger import get_logger
logger = get_logger(log_level=logging.CRITICAL)
logger.setLevel(logging.CRITICAL)
os.environ["MODELSCOPE_CACHE"] = "./"
inference_pipeline = pipeline(
task=Tasks.auto_speech_recognition,
model='damo/speech_paraformer_asr_nat-zh-cn-16k-common-vocab8404-online',
model_revision='v1.0.4'
)
model_dir = os.path.join(os.environ["MODELSCOPE_CACHE"], "damo/speech_paraformer_asr_nat-zh-cn-16k-common-vocab8404-online")
speech, sample_rate = soundfile.read(os.path.join(model_dir, "example/asr_example.wav"))
speech_length = speech.shape[0]
sample_offset = 0
chunk_size = [8, 8, 4] #[5, 10, 5] 600ms, [8, 8, 4] 480ms
stride_size = chunk_size[1] * 960
param_dict = {"cache": dict(), "is_final": False, "chunk_size": chunk_size}
final_result = ""
for sample_offset in range(0, speech_length, min(stride_size, speech_length - sample_offset)):
if sample_offset + stride_size >= speech_length - 1:
stride_size = speech_length - sample_offset
param_dict["is_final"] = True
rec_result = inference_pipeline(audio_in=speech[sample_offset: sample_offset + stride_size],
param_dict=param_dict)
if len(rec_result) != 0:
final_result += rec_result['text']
print(rec_result)
print(final_result.strip())

13
funasr/bin/asr_infer.py
View File

@ -828,9 +828,16 @@ class Speech2TextParaformerOnline:
# Change integer-ids to tokens
token = self.converter.ids2tokens(token_int)
token = " ".join(token)
results.append(token)
postprocessed_result = ""
for item in token:
if item.endswith('@@'):
postprocessed_result += item[:-2]
elif re.match('^[a-zA-Z]+$', item):
postprocessed_result += item + " "
else:
postprocessed_result += item
results.append(postprocessed_result)
# assert check_return_type(results)
return results