breizhn / DTLN-aec

This Repostory contains the pretrained DTLN-aec model for real-time acoustic echo cancellation.

breizhn/DTLN-aec Issues

The model based on 'DTLN' project and your paper about 'DTLN-aec' can't be trained as good as your pretrained one.
Updated 8 months ago5
Is this a mistake in the paper? or Am i misunderstood it ?
Closed 4 years ago2
How to training DTLN-aec model
Updated a year ago
Some questions about concatenate operation in DTLN-aec model?
Updated 2 years ago
Problem in duplicating the training process
Updated 2 years ago11
Does DTLN-aec also contain the noise suppression?
Updated 2 years ago
Can you open source your crowdsourced test data and results?
Closed 2 years ago
Just Questions this time :)
Updated 2 years ago
about repository construction plan
Updated 2 years ago
H
Closed 2 years ago
can it be used in real time voice?
Updated 3 years ago4
About the training target: nearend speech
Updated 3 years ago
Could DTLN-aec process audio frame real-time on mobile?
Closed 4 years ago2
How to use the model to generate the echo cancelled file.
Updated 3 years ago
Looking forward to more updates about training details.
Updated 3 years ago
the SNR loss is inf or nan in the training stage
Updated 3 years ago
up-sampling the phone data to 16k and then down-sampling to 8k Model effect will be better
Updated 3 years ago2
Training problems in double talk situations
Updated 3 years ago1
Can I get some files for quantization of the model?
Updated 3 years ago
Keras models needed
Updated 3 years ago
speech is totolly removed.
Closed 4 years ago9
Making Echo Cancellation working on real time voice
Closed 3 years ago
MODEL FOR aec
Updated 3 years ago
The influence of the hyperparameter encoder_size
Updated 3 years ago1
Failed on mobile device
Updated 3 years ago2
Loss decrease for a few and sudden explodes to inf
Updated 3 years ago2
where aru your training data for this project?
Closed 4 years ago