MTTS下載MTTS源代碼下載

MTTS

Ai源碼

1.0.0

下載

本項目已停止維護，已相當老舊

A Demo of MTTS Mandarin/Chinese Text to Speech FrontEnd

Mandarin/Chinese Text to Speech based on statistical parametric speech synthesis using merlin toolkit

這只是一個語音合成前端的Demo，沒有提供文本正則化，韻律預測功能，文字轉拼音使用pypinyin，分詞使用結巴分詞，這兩者的準確度也達不到商用水平。

其他語音合成項目傳送門，端到端是不錯的方向，自然度要優於merlin。

This is only a demo of mandarin frontend which is lack of some parts like "text normalization" and "prosody prediction", and the phone set && Question Set this project use havn't fully tested yet.

一個粗略的文檔：A draft documentation written in Mandarin

Data

There is no open-source mandarin speech synthesis dataset on the internet, this proj used thchs30 dataset to demostrate speech synthesis

UPDATE

open-source mandarin speech synthesis data from data-banker company, 開源的中文語音合成數據，感謝標貝公司

【數據下載】https://weixinxcxdb.oss-cn-beijing.aliyuncs.com/gwYinPinKu/BZNSYP.rar 【數據說明】http://www.data-baker.com/open_source.html

Generated Samples

Listen to https://jackiexiao.github.io/MTTS/

How To Reproduce

First, you need data contain wav and txt (prosody mark is optional)
Second, generate HTS label using this project
Using merlin/egs/mandarin_voice to train and generate Mandarin Voice

Context related annotation & Question Set

Context related annotation上下文相關標註
Question Set問題集
Rules to design a Question Set問題集設計規則

Install

Python : python3.6
System: linux(tested on ubuntu16.04)

 pip install jieba pypinyin
sudo apt-get install libatlas3-base

Run bash tools/install_mtts.sh
Or download file by yourself

Download montreal-forced-aligner and unzip to directory tools/
Download acoustic_model thchs30.zip and copy to directory misc/

Run Demo

 bash run_demo.sh

Usage

1. Generate HTS Label by wav and text

Usage: Run python src/mtts.py txtfile wav_directory_path output_directory_path (Absolute path or relative path) Then you will get HTS label, if you have your own acoustic model trained by monthreal-forced-aligner, add -a your_acoustic_model.zip , otherwise, this project use thchs30.zip acoustic model as default
Attention: Currently only support Chinese Character, txt should not have any Arabia number or English alphabet(不可包含阿拉伯數字和英文字符)

txtfile example

 A_01 这是一段文本
A_02 这是第二段文本

wav_directory example (Sampleing Rate should larger than 16khz)

 A_01.wav  
A_02.wav

2. Generate HTS Label by text with or without alignment file

Usage: Run python src/mandarin_frontend.py txtfile output_directory_path
or import mandarin_frontend

 from mandarin_frontend import txt2label

result = txt2label('向香港特别行政区同胞澳门和台湾同胞海外侨胞')
[print(line) for line in result]

# with prosody mark and alignment file (sfs file)
# result = txt2label('向#1香港#2特别#1行政区#1同胞#4澳门#2和#1台湾#1同胞#4海外#1侨胞',
            sfsfile='example_file/example.sfs')

see source code for more information, but pay attention to the alignment file(sfs file), the format is endtime phone_type not start_time, phone_type (which is different from speech ocean's data)

3. Forced-alignment

This project use Montreal-Forced-Aligner to do forced alignment, if you want to get a better alignment, use your data to train a alignment-model, see mfa: algin-using-only-the-dataset

We trained the acoustic model using thchs30 dataset, see misc/thchs30.zip , the dictionary we use mandarin_mtts.lexicon. If you use larger dataset than thchs30, you may get better alignment.
If you want to use mfa's (montreal-forced-aligner) pre-trained mandarin model, this is the dictionary you need mandarin-for-montreal-forced-aligner-pre-trained-model.lexicon