conformer and nextformer
The Nextformer implementation builds upon and extends the Conformer architecture by augmenting it with ConvNeXt components, making it a direct successor rather than a competitor or complement to the base Conformer implementation.
About conformer
sooftware/conformer
[Unofficial] PyTorch implementation of "Conformer: Convolution-augmented Transformer for Speech Recognition" (INTERSPEECH 2020)
Combines convolution layers and multi-head attention to capture both local acoustic features and long-range dependencies in a parameter-efficient architecture. Integrates with CTC loss for end-to-end training on speech recognition tasks, with model-only code designed for use in larger training frameworks like OpenSpeech.
About nextformer
tuanio/nextformer
PyTorch implementation of "Nextformer: A ConvNeXt Augmented Conformer For End-To-End Speech Recognition"
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work