Songxiang Liu's Projects
g2p: English Grapheme To Phoneme Conversion
Code for reproducing results in "Glow: Generative Flow with Invertible 1x1 Convolutions"
A Generative Flow for Text-to-Speech via Monotonic Alignment Search
Pytorch: Viterbi, Forward-Backward and Baum Welch with a Hidden Markov Model (HMM)
HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis
Joint Dictionary Learning-based Non-Negative Matrix Factorization for Voice Conversion
Python functions for reading kaldi data formats. Useful for rapid prototyping with python.
Alignment files of LibriTTS.
Personal homepage:
Efficient neural speech synthesis
This is now the official location of the Merlin project.
Simple MFCC extractor and an speech recognition algorithm (Dynamic Time Warping)
Command line utility for forced alignment using Kaldi
A Demo of Mandarin/Chinese TTS frontend
Implementation code of non-parallel sequence-to-sequence VC
Custom Cayman is a Jekyll theme for GitHub Pages
Unofficial Parallel WaveGAN (+ MelGAN & Multi-band MelGAN) with Pytorch
Praat in Python, the Pythonic way
Code accompanying ML4MD ICML 2020 paper - "Generative Modelling for Controllable Audio Synthesis of Expressive Piano Performance".
My vim comfiguration
PPG-Based Voice Conversion
All Algorithms implemented in Python
Python - 100ๅคฉไปๆฐๆๅฐๅคงๅธ
pytorch-kaldi is a project for developing state-of-the-art DNN/RNN hybrid speech recognition systems. The DNN part is managed by pytorch, while feature extraction, label computation, and decoding are performed with the kaldi toolkit.
My personal homepage
A python package to analyze and compare voices with deep learning