Fabian-Robert Stöter

Train single channel model using left or right channel

2

Currently using `train.py --nb-channels 1` will apply a downmix in the spectral domain inside the model to feed in only single channel audio. However, I can think about applications where...

Add SI-SDR

12

Add scale invariant SDR metric. See https://arxiv.org/abs/1811.02508

Refactoring

1

* [ ] reduce the number of functions * [ ] add modules for different bsseval versions and other metrics such as SNR, and SI-SDR * [ ] possible look...

* [ ] use museval as a test dependency to load the museval json files * [ ] later extend museval so that the json files include the used metrics...

With more and more libs using museval for evaluation, writing the museval json files, it would be great to have cli command to just print the aggregated metrics in the...

Implementation Status

1

The nnabla release of open-unmix is not feature complete with respect to our pytorch reference. The following issues would need to be solved * [X] Dataset parameters * [ ]...

python wrapper

1

I just found this project. This is great... Before I try it myself: do you have experience interfacing gstpeaq with [gst-python](https://github.com/GStreamer/gst-python)?

Add ability to use `from_ffmpeg` as GraphIOTensor

5

the audio graph modes as introduced in #615 greatly improves audio loading in `tf.data` pipelines. When external (ffmpeg) decoding is required (in addition to #650), it would be great if...