Download CMake from: https://cmake.org/download/
wget https://cmake.org/files/v3.12/cmake-3.12.3.tar.gz
tar zxvf cmake-3.*
wget https://cmake.org/files/v3.12/cmake-3.12.3.tar.gz
tar zxvf cmake-3.*
#!/usr/bin/env python | |
# -*- coding:UTF-8 -*- | |
import torch | |
import torch.nn as nn | |
import torch.nn.init as init | |
def weight_init(m): | |
''' |
################################################################################ | |
# Method 1: Install using rpm packages (credit to DarkMukke) | |
# | |
rpm -Uvh http://mirror.ghettoforge.org/distributions/gf/gf-release-latest.gf.el7.noarch.rpm | |
rpm --import http://mirror.ghettoforge.org/distributions/gf/RPM-GPG-KEY-gf.el7 | |
# WARNING: removing vim-minimal uninstalls `sudo` if you skip the second step | |
# make sure to at least run `yum install sudo` | |
yum -y remove vim-minimal vim-common vim-enhanced |
__author__ = 'Christoph Heindl' | |
__copyright__ = 'Copyright 2017' | |
__license__ = 'BSD' | |
"""Trains a HMM based on gradient descent optimization. | |
The parameters (theta) of the model are transition and | |
emission probabilities, as well as the initial state probabilities. | |
Given a start solution, the negative log likelihood of data given the |
import numpy as np | |
import tensorflow as tf | |
from tensorflow.python.layers import core as layers_core | |
hparams = tf.contrib.training.HParams( | |
batch_size=3, | |
encoder_length=4, | |
decoder_length=5, | |
num_units=6, | |
src_vocab_size=7, |
本文综述近年来深度学习用于语音合成的一些方法。
在自回归生成模型在图像和文本领域广泛应用的时候,WaveNet [4] 尝试将这些思想应用于语音领域。仿照PixelRNN (van den Oord et al., 2016)图像生成的做法, WaveNet依据之前采样点来生成下一个采样点。生成下一个采样点的模型为CNN结构。为了生成指定说话人的声音,以及生成指定文本的声音,引入了全局条件和局部条件,来控制合成内容。为了扩大感受野,带洞卷积,使filter的按照指数扩张。
WaveNet存在的问题是,1) 每次预测一个采样点,速度太慢;2)如果用于TTS,那初始采样点选择将会很重要;3)以及需要文本前端的支持,前端分析出错,将直接影响合成效果。
"""Script to illustrate usage of tf.estimator.Estimator in TF v1.3""" | |
import tensorflow as tf | |
from tensorflow.examples.tutorials.mnist import input_data as mnist_data | |
from tensorflow.contrib import slim | |
from tensorflow.contrib.learn import ModeKeys | |
from tensorflow.contrib.learn import learn_runner | |
# Show debugging output |
""" | |
Author: Awni Hannun | |
This is an example CTC decoder written in Python. The code is | |
intended to be a simple example and is not designed to be | |
especially efficient. | |
The algorithm is a prefix beam search for a model trained | |
with the CTC loss function. |
def _sequence_mask(sequence_length, max_len=None): | |
if max_len is None: | |
max_len = sequence_length.data.max() | |
batch_size = sequence_length.size(0) | |
seq_range = torch.range(0, max_len - 1).long() | |
seq_range_expand = seq_range.unsqueeze(0).expand(batch_size, max_len) | |
seq_range_expand = Variable(seq_range_expand) | |
if sequence_length.is_cuda: | |
seq_range_expand = seq_range_expand.cuda() | |
seq_length_expand = (sequence_length.unsqueeze(1) |
yum install -y boost boost-devel | |
yum install -y make gcc gcc-c++ kernel-devel python-devel | |
wget https://github.com/arvidn/libtorrent/releases/download/libtorrent-1_0_10/libtorrent-rasterbar-1.0.10.tar.gz | |
tar zxvf libtorrent-rasterbar-1.0.10.tar.gz | |
cd libtorrent-rasterbar-1.0.10.tar.gz | |
./configure --disable-debug --with-boost-libdir=/usr/lib64 --disable-encryption --enable-python-binding | |
make && make install | |
export LD_LIBRARY_PATH=/usr/local/lib/ | |
cd bindings/python | |
python setup.py build |