lucidrains/conformer
PyTorch implementation of the Conformer architecture combining convolution modules with Transformer self-attention for speech recognition.
Not currently ranked — collecting fresh signals.
star history
This repository provides a PyTorch implementation of the Conformer model, which augments Transformer architectures with convolutional modules to improve local inductive bias. The implementation includes the convolutional module, individual Conformer blocks, and full multi-block Conformer models with configurable dimensions, attention heads, and feed-forward multipliers. It is designed for sequence-to-sequence tasks, particularly automatic speech recognition.
Frequently asked
- What is lucidrains/conformer?
- PyTorch implementation of the Conformer architecture combining convolution modules with Transformer self-attention for speech recognition.
- Is conformer open source?
- Yes — lucidrains/conformer is open source, released under the MIT license.
- What language is conformer written in?
- lucidrains/conformer is primarily written in Python.
- How popular is conformer?
- lucidrains/conformer has 437 stars on GitHub.
- Where can I find conformer?
- lucidrains/conformer is on GitHub at https://github.com/lucidrains/conformer.