Resample training input samples to align with inference restraints

[alchemi5t]

Continuing the discussion from:

Resampling to 16khz

@lissyx

There could be value in ensuring that training can be done at other sample rates than 16kHz, but I’m unsure that resampling is the proper solution, to be honest.

I have added a variable(can be changed to a flag) which can be set to desired target SR. But won’t having a single sample rate(same as training and inference) improve convergence?

I’ve tested training models on different SR and inferencing on 16khz, and as expected, the model produces unacceptable results(WER, CER, LOSS and output put together). But the same model infers much better when using the same SR test files(Test results after training).

It does retain the original audio characteristics after resampling.

Do you think this would not be useful for training?

[alchemi5t]

Maybe resample it to a bracket near 16khz (E.g., 14-18 khz), if you think other SRs might be worth preserving(If only for having different SRs in the dataset for helping generalize the model).

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

But won’t having a single sample rate(same as training and inference) improve convergence?

I don’t get your point: this is exactly the current situation, training at 16kHz, inference at 16kHz.

[alchemi5t]

The point being i can train at 44khz and not infer on it.

Trained model on my own data

@lucifera678 ok ok listen here. go to util/flags.py . change audio_sample_rate to 16000(you set it as 44100). and then you’ll see that you can export your model. Do i know if that screws up your model? I do not. But can you export it? yes. good luck.

Like this guy.

Instead of him having to do whatever he did, if the code resampled his 44khz data to 16k, then he would have not had that issue.

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

The point being i can train at 44khz and not infer on it.

I don’t understand a single word of what you say. Each sentence contradicts the next one.

[alchemi5t]

I meant, I can train a model on data with 44khz but my inference will require 16khz data which would make it hard for the model to predict accurately, if that makes sense.

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

I still don’t get what you are trying to achieve. Can you describe precisely your problem ?

[alchemi5t]

let me simplify the issue. I want to know if i can run inference on any other SR other than 16k,ATM without code modification.

[reuben]

Yes, just specify the sample rate when passing in the audio. Every function that takes audio samples in the API also takes a sample rate.

[alchemi5t]

Got it. I thought only 16k was possible(misinterpreted the readme.). Thanks for clearing things up.