Skip to content

Commit

Permalink
Update README.md
Browse files Browse the repository at this point in the history
  • Loading branch information
hongwen-sun committed Sep 27, 2018
1 parent 0fae5ce commit 2991ec4
Showing 1 changed file with 1 addition and 1 deletion.
2 changes: 1 addition & 1 deletion README.md
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,7 @@
### 基于python的中文语音识别系统.
包含声学模型和语言模型两个部分组成,两个模型都是基于神经网络。

- 声学模型 - `acoustic_model`文件夹下,**如果想直接运行,建议直接运行**`cnn_with_full.py`
- 声学模型 - `acoustic_model`文件夹下,**建议直接运行**`cnn_with_full.py`
- 该项目实现了GRU-CTC中文语音识别声音模型,所有代码都在`gru_ctc_am.py`中,包括:
- 增加了基于科大讯飞DFCNN的CNN-CTC结构的中文语音识别模型`cnn_ctc_am.py`,与GRU相比,对网络结构进行了稍加改造。
- 完全使用DFCNN框架搭建声学模型,稍加改动,将部分卷积层改为inception,使用时频图作为输入,`cnn_with_fbank.py`
Expand Down

0 comments on commit 2991ec4

Please sign in to comment.