Incredible! How to run inference on a custom file?

[youssefavx]

Super impressed by your results! Curious to know how I could run a sample audio file through your model to upsample it. It seems the code provided here simply evaluates the model: https://github.com/haoheliu/ssr_eval/tree/main/examples/NVSR

I'll try to figure it out from that but would love any help whatsoever. No pressure whatsoever if busy though!

[youssefavx]

Also does this require fine-tuning on a custom voice for good results?

[youssefavx]

Okay I think I figured it out, please let me know if I'm using it incorrectly though:

import soundfile as sf

if(torch.cuda.is_available()): device = "cuda"
else: device="cpu"

testee = NVSRPostProcTestee(device)
x, _ = librosa.load("Sample.wav", sr=44100)
result = testee.infer(x)
sf.write('result.wav', result, 44100)

[haoheliu]

Yes, that's the way it works :)

Okay I think I figured it out, please let me know if I'm using it incorrectly though:

import soundfile as sf

if(torch.cuda.is_available()): device = "cuda"
else: device="cpu"

testee = NVSRPostProcTestee(device)
x, _ = librosa.load("Sample.wav", sr=44100)
result = testee.infer(x)
sf.write('result.wav', result, 44100)

Yes, that's how it works :)