markovka17/dla
A university course on deep learning for audio processing covering speech recognition, TTS, speaker verification, and source separation.
Not currently ranked — collecting fresh signals.
star history
An educational course on deep learning for audio, taught at HSE CS Faculty. The course covers digital signal processing fundamentals, automatic speech recognition (CTC, LAS, RNN-T), source separation, text-to-speech, voice conversion, and audio self-supervised learning. Materials include weekly lectures and seminars with hands-on practice using PyTorch.
Frequently asked
- What is markovka17/dla?
- A university course on deep learning for audio processing covering speech recognition, TTS, speaker verification, and source separation.
- Is dla open source?
- Yes — markovka17/dla is open source, released under the MIT license.
- What language is dla written in?
- markovka17/dla is primarily written in Jupyter Notebook.
- How popular is dla?
- markovka17/dla has 755 stars on GitHub.
- Where can I find dla?
- markovka17/dla is on GitHub at https://github.com/markovka17/dla.