buriburisuri/speech-to-text-wavenet
A TensorFlow implementation adapting DeepMind's WaveNet generative audio model for end-to-end sentence-level English speech recognition.
Not currently ranked — collecting fresh signals.
star history
This repository implements an end-to-end speech recognition system based on DeepMind’s WaveNet architecture. It uses dilated convolutions and Connectionist Temporal Classification (CTC) loss to train on raw audio features extracted as MFCC from the VCTK dataset. The model processes speech audio to produce text transcriptions without intermediate phoneme stages.
Frequently asked
- What is buriburisuri/speech-to-text-wavenet?
- A TensorFlow implementation adapting DeepMind's WaveNet generative audio model for end-to-end sentence-level English speech recognition.
- Is speech-to-text-wavenet open source?
- Yes — buriburisuri/speech-to-text-wavenet is open source, released under the Apache-2.0 license.
- What language is speech-to-text-wavenet written in?
- buriburisuri/speech-to-text-wavenet is primarily written in Python.
- How popular is speech-to-text-wavenet?
- buriburisuri/speech-to-text-wavenet has 4k stars on GitHub.
- Where can I find speech-to-text-wavenet?
- buriburisuri/speech-to-text-wavenet is on GitHub at https://github.com/buriburisuri/speech-to-text-wavenet.