search

breizhn / DTLN-aec

This Repostory contains the pretrained DTLN-aec model for real-time acoustic echo cancellation.
MIT License
280 stars 73 forks source link

can it be used in real time voice? #9

Open dttlgotv opened 3 years ago

dttlgotv commented 3 years ago

It seems that it takes long time to process data after I tried you demo. Can it use in real time voice communication? or is it only used in offline voice process.

breizhn commented 3 years ago

Hi!

From its design it can be used in real-time voice communication. For small devices the large model is probably too big. But it can probably easily run when offloading the inference to a neural network engine (when available).

pandysong commented 3 years ago

The model looks not so big. Is it possible to run the model on the embedded SoC with NPU (Neural Process Unit) which usually process images (with quantization)?

Every much interested in running this on a SoC if possible.

StuartIanNaylor commented 2 years ago

Have a look at what sanebow did https://github.com/SaneBow/PiDTLN

pandysong commented 2 years ago

Have a look at what sanebow did https://github.com/SaneBow/PiDTLN

@StuartIanNaylor Thank you. I will check.