Christoph Boeddeker

When we added pesq, we considered three implementations: - Our own implementation that used subprosess - `+` Zero modifications of original code - `-` Need file system - `+` Can...

I was hoping that https://github.com/ludlows/python-pesq fixes this problem. It looks like a huge modification of the source code is necessary and I don't know when they fix it. It is...

As far as I know, your pesq code is thread safe, because the python GIL is not released. Letting the code release the GIL may be difficult, because the original...

Yes, subprocess allows to execute the computation in parallel. An another alternative is multiprocessing. I personally use most of the time MPI, where I calculate PESQ, there I don't care...

Hello, this error message tells you, that the shape of the initialization is wrong. It should have 1165 frames to match the observation, but has only 1140 frames. My guess...

Dear zcy618, do you mean, how you can run the file https://github.com/fgnt/pb_bss/blob/master/examples/mixture_model_example.ipynb ? This is a jupyter notebook. In the documentation for jupyter, you can read, how to launch a...

I am not sure, what you mean with inference. The mixture models are categorized as blind source separation technique. They have no training phase like neuronal networks. They learn the...

> Can this be used with a microphone array for example to do source separation? Yes, the idea of the mixture models (cACG and complex Watson) is to model spatial...

> So if all i have is the observation, i.e. the multichannel signal, what are speech_image_0.wav, speech_image_1.wav, ..., noise_image.wav, speech_source_0.wav, speech_source_1.wav, ... ? These files were used to generate the...

The `OraclePermutationAlignment` is used to identify the first speaker, the second speaker and the noise. In a real application you cannot do this. Instead, you need to identify the noise....