Skip to content
RNN CTC by using TensorFlow.
Branch: master
Clone or download
ugnelis Merge pull request #16 from ugnelis/multi-layers-fix
Fixed multi layers issue. Now it's possible to have multiple layers.
Latest commit 8fc5457 Jan 30, 2018
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
data/LibriSpeech Added example data. Added text file and audio file reading. Sep 28, 2017
tests
.gitignore
.travis.yml Add initial test, fix update travis config Oct 12, 2017
CONTRIBUTING.md
LICENSE Added license. Oct 2, 2017
README.md
demo.py
requirements.txt add requirements.txt Oct 4, 2017
train.py
utils.py

README.md

TensorFlow RNN CTC Build Status

Connectionist Temporal Classification (CTC) by using Recurrent Neural Network (RNN) in TensorFlow.

Requirements

  • Python 2.7+ (for Linux)
  • Python 3.5+ (for Windows)
  • TensorFlow 1.0+
  • NumPy 1.5+
  • SciPy 0.12+
  • python_speech_features 0.1+

Installation

I suggest you to use Anaconda. For TensorFlow and python_speech_features use pip:

$ activate anaconda_env_name
(anaconda_env_name)$ pip install python_speech_features
(anaconda_env_name)$ pip install --ignore-installed --upgrade tensorflow # without GPU
(anaconda_env_name)$ pip install --ignore-installed --upgrade tensorflow-gpu # with GPU

Training

Run training by using train.py file.

python train.py

License

This project is licensed under the terms of the MIT license.

You can’t perform that action at this time.
You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session.