sooftware / conformer

[Unofficial] PyTorch implementation of "Conformer: Convolution-augmented Transformer for Speech Recognition" (INTERSPEECH 2020)
Apache License 2.0
943 stars 174 forks source link

Request for Implementation of Memory-Efficient Attention in Conformer #63

Open N-damo opened 2 weeks ago

N-damo commented 2 weeks ago

Hello Conformer developers and contributors,

I have been using the Conformer model for my research and I'm very impressed with its performance. However, I noticed that the current implementation does not include memory-efficient attention, which is crucial for handling long sequences and reducing memory usage during training and inference. Best regards, LianLin