GitHub - Cosmos-Break/asr: 沪语（上海话）ASR（语音识别）模型

使用的工具库

Huggingface的transformers和huggingsound。

包含一个沪语ASR模型(沪语语音->沪语转写文本)和机器翻译模型(沪语转写文本->普通话文本)。

包含Magichub开源数据集、喜马拉雅，shh.dict.cn，中国语言网爬取的数据集、讯飞TTS生成的wav数据。

train.py 用于训练ASR模型

train_translation.py 用于训练MT模型

使用fastapi进行整个模型的部署，运行run_service.sh部署。

Name	Name	Last commit message	Last commit date
Latest commit Cosmos-Break push May 13, 2024 dca6aaa · May 13, 2024 History 61 Commits
.vscode	.vscode	push	Sep 6, 2022
Shanghai_Dialect_Conversational_Speech_Corpus	Shanghai_Dialect_Conversational_Speech_Corpus	push	Sep 6, 2022
Shanghai_Dialect_Dict	Shanghai_Dialect_Dict	push	Sep 6, 2022
Shanghai_Dialect_Scripted_Speech_Corpus_Daily_Use_Sentence	Shanghai_Dialect_Scripted_Speech_Corpus_Daily_Use_Sentence	push	Sep 6, 2022
Shanghai_Dialect_Ximalaya	Shanghai_Dialect_Ximalaya	push	Sep 9, 2022
Shanghai_Dialect_Zhongguoyuyan	Shanghai_Dialect_Zhongguoyuyan	push	Sep 13, 2022
app	app	push	Sep 6, 2022
.gitignore	.gitignore	push	Sep 2, 2022
README.md	README.md	Update README.md	Nov 9, 2022
change_vocab_size.py	change_vocab_size.py	push	May 13, 2024
collect_vocab.py	collect_vocab.py	push	May 13, 2024
make_requests.py	make_requests.py	push	Sep 6, 2022
push.sh	push.sh	push	Aug 24, 2022
requirements.txt	requirements.txt	push	Sep 13, 2023
run.sh	run.sh	push	Aug 31, 2022
run_service.sh	run_service.sh	push	Sep 6, 2022
split_audio_1.py	split_audio_1.py	push	Aug 26, 2022
split_audio_2.py	split_audio_2.py	push	Aug 26, 2022
split_audio_3.py	split_audio_3.py	push	Sep 2, 2022
split_audio_4.py	split_audio_4.py	push	Sep 9, 2022
train.py	train.py	push	Sep 13, 2023
train_translation.py	train_translation.py	push	Sep 9, 2022
transcribe.py	transcribe.py	push	Sep 13, 2023
translation.py	translation.py	push	Sep 2, 2022