OrgLion-ML
/
ASRT_SpeechRecognition
mirror of https://github.com/nl8590687/ASRT_SpeechRecognition.git
1
0
Fork 0
Code Issues Releases Wiki Activity

update readme

This commit is contained in:
nl 2020-04-18 14:35:39 +08:00
parent 4e4aad07ce
commit 20f9a9c0fb
2 changed files with 30 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
README.md
View File

@ -1,7 +1,10 @@
# A Deep-Learning-Based Chinese Speech Recognition System # A Deep-Learning-Based Chinese Speech Recognition System
基于深度学习的中文语音识别系统，如果您觉得喜欢，请点一个 **"Star"** 吧~ 基于深度学习的中文语音识别系统，如果您觉得喜欢，请点一个 **"Star"** 吧~
[![GPL-3.0 Licensed](https://img.shields.io/badge/License-GPL3.0-blue.svg?style=flat)](https://opensource.org/licenses/GPL-3.0) [![TensorFlow Version](https://img.shields.io/badge/Tensorflow-1.4+-blue.svg)](https://www.tensorflow.org/) [![Keras Version](https://img.shields.io/badge/Keras-2.0+-blue.svg)](https://keras.io/) [![Python Version](https://img.shields.io/badge/Python-3.x-blue.svg)](https://www.python.org/) [![GPL-3.0 Licensed](https://img.shields.io/badge/License-GPL3.0-blue.svg?style=flat)](https://opensource.org/licenses/GPL-3.0)
[![TensorFlow Version](https://img.shields.io/badge/Tensorflow-1.13+-blue.svg)](https://www.tensorflow.org/)
[![Keras Version](https://img.shields.io/badge/Keras-2.2+-blue.svg)](https://keras.io/)
[![Python Version](https://img.shields.io/badge/Python-3.x-blue.svg)](https://www.python.org/)
**ReadMe Language** | 中文版 | [English](https://github.com/nl8590687/ASRT_SpeechRecognition/blob/master/README_EN.md) | **ReadMe Language** | 中文版 | [English](https://github.com/nl8590687/ASRT_SpeechRecognition/blob/master/README_EN.md) |
@ -19,6 +22,12 @@ ASRT的原理请查看本文：
* [统计语言模型：从中文拼音到文本](https://blog.ailemon.me/2017/04/27/statistical-language-model-chinese-pinyin-to-words/) * [统计语言模型：从中文拼音到文本](https://blog.ailemon.me/2017/04/27/statistical-language-model-chinese-pinyin-to-words/)
* [无需中文分词算法的简单词频统计](https://blog.ailemon.me/2017/02/20/simple-words-frequency-statistic-without-segmentation-algorithm/) * [无需中文分词算法的简单词频统计](https://blog.ailemon.me/2017/02/20/simple-words-frequency-statistic-without-segmentation-algorithm/)
关于CTC的问题请看：
* [[翻译]使用CTC进行序列建模](<https://blog.ailemon.me/2019/07/18/sequence-modeling-with-ctc/>)
更多内容请访问作者的博客：[AI柠檬博客](https://blog.ailemon.me/)
## Introduction 简介 ## Introduction 简介
本项目使用Keras、TensorFlow基于深度卷积神经网络和长短时记忆神经网络、注意力机制以及CTC实现。 本项目使用Keras、TensorFlow基于深度卷积神经网络和长短时记忆神经网络、注意力机制以及CTC实现。
@ -102,8 +111,8 @@ CNN + LSTM/GRU + CTC
Python的依赖库 Python的依赖库
* python_speech_features * python_speech_features
* TensorFlow * TensorFlow (1.13+)
* Keras * Keras (2.2+)
* Numpy * Numpy
* wave * wave
* matplotlib * matplotlib
@ -116,6 +125,9 @@ Python的依赖库
[程序运行依赖环境详细说明](https://github.com/nl8590687/ASRT_SpeechRecognition/wiki/Dependent-Environment) [程序运行依赖环境详细说明](https://github.com/nl8590687/ASRT_SpeechRecognition/wiki/Dependent-Environment)
## Data Sets 数据集 ## Data Sets 数据集
[几个最新免费开源的中文语音数据集](https://blog.ailemon.me/2018/11/21/free-open-source-chinese-speech-datasets/)
* **清华大学THCHS30中文语音数据集** * **清华大学THCHS30中文语音数据集**
data_thchs30.tgz data_thchs30.tgz

18
README_EN.md
View File

@ -1,6 +1,9 @@
# A Deep-Learning-Based Chinese Speech Recognition System # A Deep-Learning-Based Chinese Speech Recognition System
[![GPL-3.0 Licensed](https://img.shields.io/badge/License-GPL3.0-blue.svg?style=flat)](https://opensource.org/licenses/GPL-3.0) [![TensorFlow Version](https://img.shields.io/badge/Tensorflow-1.4+-blue.svg)](https://www.tensorflow.org/) [![Keras Version](https://img.shields.io/badge/Keras-2.0+-blue.svg)](https://keras.io/) [![Python Version](https://img.shields.io/badge/Python-3.x-blue.svg)](https://www.python.org/) [![GPL-3.0 Licensed](https://img.shields.io/badge/License-GPL3.0-blue.svg?style=flat)](https://opensource.org/licenses/GPL-3.0)
[![TensorFlow Version](https://img.shields.io/badge/Tensorflow-1.13+-blue.svg)](https://www.tensorflow.org/)
[![Keras Version](https://img.shields.io/badge/Keras-2.2+-blue.svg)](https://keras.io/)
[![Python Version](https://img.shields.io/badge/Python-3.x-blue.svg)](https://www.python.org/)
**ReadMe Language** | [中文版](https://github.com/nl8590687/ASRT_SpeechRecognition/blob/master/README.md) | English | **ReadMe Language** | [中文版](https://github.com/nl8590687/ASRT_SpeechRecognition/blob/master/README.md) | English |
@ -17,6 +20,12 @@ For questions about the principles of the statistical language model that are of
* [Simple word frequency statistics without Chinese word segmentation algorithm (Chinese)](https://blog.ailemon.me/2017/02/20/simple-words-frequency-statistic-without-segmentation-algorithm/) * [Simple word frequency statistics without Chinese word segmentation algorithm (Chinese)](https://blog.ailemon.me/2017/02/20/simple-words-frequency-statistic-without-segmentation-algorithm/)
* [Statistical Language Model: Chinese Pinyin to Words (Chinese)](https://blog.ailemon.me/2017/04/27/statistical-language-model-chinese-pinyin-to-words/) * [Statistical Language Model: Chinese Pinyin to Words (Chinese)](https://blog.ailemon.me/2017/04/27/statistical-language-model-chinese-pinyin-to-words/)
For questions about CTC, see:
* [[Translation] Sequence Modeling with CTC (Chinese)](<https://blog.ailemon.me/2019/07/18/sequence-modeling-with-ctc/>)
For more infomation please refer to author's blog website: [AILemon Blog](https://blog.ailemon.me/) (Chinese)
## Introduction ## Introduction
This project uses Keras, TensorFlow based on deep convolutional neural network and long-short memory neural network, attention mechanism and CTC to implement. This project uses Keras, TensorFlow based on deep convolutional neural network and long-short memory neural network, attention mechanism and CTC to implement.
@ -102,8 +111,8 @@ However, as the current international and domestic teams can achieve 98%, the ac
## Python libraries that need importing ## Python libraries that need importing
* python_speech_features * python_speech_features
* TensorFlow * TensorFlow (1.13+)
* Keras * Keras (2.2+)
* Numpy * Numpy
* wave * wave
* matplotlib * matplotlib
@ -116,6 +125,9 @@ However, as the current international and domestic teams can achieve 98%, the ac
[Dependent Environment Details](https://github.com/nl8590687/ASRT_SpeechRecognition/wiki/Dependent-Environment) [Dependent Environment Details](https://github.com/nl8590687/ASRT_SpeechRecognition/wiki/Dependent-Environment)
## Data Sets ## Data Sets
[Some free Chinese speech datasets (Chinese)](https://blog.ailemon.me/2018/11/21/free-open-source-chinese-speech-datasets/)
* **Tsinghua University THCHS30 Chinese voice data set** * **Tsinghua University THCHS30 Chinese voice data set**
data_thchs30.tgz data_thchs30.tgz