资源简介
一个Text to speech开源框架,
代码片段和文件信息
import numpy as np
import tensorflow as tf
# Default hyperparameters
hparams = tf.contrib.training.HParams(
# Comma-separated list of cleaners to run on text prior to training and eval. For non-English
# text you may want to use “basic_cleaners“ or “transliteration_cleaners“.
cleaners=‘english_cleaners‘
###########################################################################################################################################
#Audio
num_mels = 80 #Number of mel-spectrogram channels and local conditioning dimensionality
num_freq = 1025 # (= n_fft / 2 + 1) only used when adding linear spectrograms post processing network
rescale = True #Whether to rescale audio prior to preprocessing
rescaling_max = 0.999 #Rescaling value
trim_silence = True #Whether to clip silence in Audio (at beginning and end of audio only not the middle)
clip_mels_length = True #For cases of OOM (Not really recommended working on a workaround)
max_mel_frames = 1300 #Only relevant when clip_mels_length = True
# Use LWS (https://github.com/Jonathan-LeRoux/lws) for STFT and phase reconstruction
# It‘s preferred to set True to use with https://github.com/r9y9/wavenet_vocoder
# Does not work if n_ffit is not multiple of hop_size!!
use_lws=False
silence_threshold=2 #silence threshold used for sound trimming for wavenet preprocessing
#Mel spectrogram
n_fft = 2048 #Extra window size is filled with 0 paddings to match this parameter
hop_size = 300 #For 22050Hz 275 ~= 12.5 ms
win_size = 1200 #For 22050Hz 1100 ~= 50 ms (If None win_size = n_fft)
sample_rate = 24000 #22050 Hz (corresponding to ljspeech dataset)
frame_shift_ms = None
#M-AILABS (and other datasets) trim params
trim_fft_size = 512
trim_hop_size = 128
trim_top_db = 23
#Mel and Linear spectrograms normalization/scaling and clipping
signal_normalization = True
allow_clipping_in_normalization = True #Only relevant if mel_normalization = True
symmetric_mels = False #Whether to scale the data to be symmetric around 0
max_abs_value = 4. #max absolute value of data. If symmetric data will be [-max max] else [0 max]
normalize_for_wavenet = True #whether to rescale to [0 1] for wavenet.
#Limits
min_level_db = -100
ref_level_db = 20
fmin = 0 #Set this to 75 if your speaker is male! if female 125 should help taking off noise. (To test depending on dataset)
fmax = 7600
#Griffin Lim
power = 1.5
griffin_lim_iters = 60
###########################################################################################################################################
#Tacotron
outputs_per_step = 2 #number of frames to generate at each decoding step (speeds up computation and allows for higher batch size)
stop_at_any = True #Determines whether the decoder should stop when predicting to any frame or to all of them
embedding_dim = 512 #dimension of embedding space
enc_conv_num_layers = 3 #number of encoder convolutional layers
enc
属性 大小 日期 时间 名称
----------- --------- ---------- ----- ----
目录 0 2020-08-19 07:15 Tacotron-2-master\
文件 212 2020-08-19 07:15 __MACOSX\._Tacotron-2-master
文件 8196 2020-08-19 07:15 Tacotron-2-master\.DS_Store
文件 120 2020-08-19 07:15 __MACOSX\Tacotron-2-master\._.DS_Store
文件 1250 2018-08-14 19:22 Tacotron-2-master\.gitignore
文件 212 2018-08-14 19:22 __MACOSX\Tacotron-2-master\._.gitignore
目录 0 2018-10-23 09:31 Tacotron-2-master\.idea\
目录 0 2018-09-03 08:20 Tacotron-2-master\.ipynb_checkpoints\
目录 0 2018-10-23 08:13 Tacotron-2-master\datasets\
文件 212 2018-10-23 08:13 __MACOSX\Tacotron-2-master\._datasets
文件 1540 2018-10-23 08:33 Tacotron-2-master\griffin_lim_synthesis_tool.ipynb
文件 212 2018-10-23 08:33 __MACOSX\Tacotron-2-master\._griffin_lim_synthesis_tool.ipynb
文件 14285 2018-08-14 19:22 Tacotron-2-master\hparams.py
文件 212 2018-08-14 19:22 __MACOSX\Tacotron-2-master\._hparams.py
文件 1260 2018-08-14 19:22 Tacotron-2-master\infolog.py
文件 212 2018-08-14 19:22 __MACOSX\Tacotron-2-master\._infolog.py
文件 1069 2018-08-14 19:22 Tacotron-2-master\LICENSE
文件 212 2018-08-14 19:22 __MACOSX\Tacotron-2-master\._LICENSE
目录 0 2018-08-14 19:22 Tacotron-2-master\papers\
文件 212 2018-08-14 19:22 __MACOSX\Tacotron-2-master\._papers
文件 4182 2018-08-14 19:22 Tacotron-2-master\preprocess.py
文件 212 2018-08-14 19:22 __MACOSX\Tacotron-2-master\._preprocess.py
文件 8691 2018-08-14 19:22 Tacotron-2-master\README.md
文件 212 2018-08-14 19:22 __MACOSX\Tacotron-2-master\._README.md
文件 204 2018-08-28 03:28 Tacotron-2-master\requirements.txt
文件 176 2018-08-28 03:28 __MACOSX\Tacotron-2-master\._requirements.txt
文件 1243 2018-08-14 19:22 Tacotron-2-master\sentences.txt
文件 268 2018-08-14 19:22 __MACOSX\Tacotron-2-master\._sentences.txt
文件 4516 2018-08-14 19:22 Tacotron-2-master\synthesize.py
文件 268 2018-08-14 19:22 __MACOSX\Tacotron-2-master\._synthesize.py
目录 0 2018-10-23 08:14 Tacotron-2-master\tacotron\
............此处省略152个文件信息
- 上一篇:python实现自动操作windows应用
- 下一篇:IIs put上传工具
相关资源
- 基于PyQt5的视频播放器设计
- python批量获取网站上的视频地址
- 网易云课堂视频(python脚本)
- python合并多个mp4视频文件成一个mp4文
- 网页视频并合并(2heiPage.py)
- 深度学习视频中的行为识别
- Python-使用DeepFakes实现YouTube视频自动换
- Mofan莫烦python全部教程代码
- 一套最新价值1680元的python爬虫实战全
- 测试工程师相关学习视频(包含pyth
- 093 2018北风网人工智能视频(完结)转
- Python-冲顶大会芝士超人西瓜视频头脑
- 中国大学MOOC课件爬取含视频
- 从视频中分离前景目标的Python & Matl
- opencv3+python人脸检测和识别- 完整实战
- opencv3+python人脸检测和识别- 完整实战
- opencv3+python人脸检测和识别 完整项目
- opencv3视频中检测人脸python
- 《Python金融序列量化应用编程指南》
- OpenCV-Python中文教程
- python3+实现视频转图片和图片转视频
- 基于python的道路视频车道线检测
- MDNet视频目标跟踪PyTorch源码Python 2.7版
- Python-指定用户的所有抖音视频以及收
- Python-西瓜视频百万英雄答题助手
- python教学视频
- Python入门教程完整版懂中文就能学会
- python基础教程第二版中文版 完整高清
- 2017年传智黑马python基础加就业班视频
- 传智博客-黑马-2017-Python就业班-完整版
评论
共有 条评论