winddori2002
winddori2002
Thank you for answering! I understand and solve it.
Hi, We have uploaded some audio pairs in the git (./samples). We randomly selected pairs and tried to choose some pairs containing more noises. Thanks.
Hi! We have uploaded the code for custom enhancement. You can find details in './custom_enhance.py'. The code is only for inference, and if you want to evaluate, you can refer...
I will upload the weights soon. Thank you.
Hi, We have uploaded both MANNER (small) and MANNER (base) a while ago. The weight you tried is MANNER (base) maybe. You can check the updated release. Thank you
Hi, I think it's ok since the speaker encoder indirectly learns to extract speaker identity. I tried other features such as wav2vec2.0, but it was less effective than CPC features....
Hi, The results are obtained by running main.py test after training. You can refer to generate_eval_pair.py to distinguish s2s and u2u. It generates metafiles for testing s2s and u2u. It...
I think it uses a similar process. You can check ./src/metric.py to get scores related to speaker verification. Thanks.
Hi, thanks for your interest. Each normalization can be used to preserve or emphasize specific aspects of these axes. IN preserves per-sample information. The result of IN retains temporal structure,...