Some more debug prints

app.py

@@ -43,6 +43,7 @@ def mix_at_snr(clean, noise, snr, eps=1e-10):
     K = torch.sqrt((E_noise / E_speech) * 10 ** (snr / 10) + eps)
     noise = noise / K
     mixture = clean + noise
+    print("mixture:", mixture.shape)
     assert torch.isfinite(mixture).all()
     max_m = mixture.abs().max()
     if max_m > 1:
@@ -51,25 +52,28 @@ def mix_at_snr(clean, noise, snr, eps=1e-10):
     return clean, noise, mixture
 
 
-def mix_and_denoise(speech, speech_alt, noise, snr):
+def mix_and_denoise(speech_rec, speech_upl, noise, snr):
     sr = config("sr", 48000, int, section="df")
-    print(speech, noise, snr)
+    print(speech_rec, noise, snr)
     if noise is None:
         noise = "samples/dkitchen.wav"
     sp_kwargs = {}
-    if speech is None or "none" in speech:
+    if speech_rec is None or "none" in speech_rec:
         speech = "samples/p232_013_clean.wav"
-        if speech_alt is not None and "none" not in speech_alt:
+        if speech_upl is not None and "none" not in speech_upl:
             print("using speech_alt")
-            speech = speech_alt
+            speech = speech_upl
     else:
+        speech = speech_rec
         sp_kwargs = {"frame_offset": 4800}
-    speech, _ = load_audio(speech, sr, **sp_kwargs)
+    speech, _ = load_audio(speech_rec, sr, **sp_kwargs)
     print(f"Loaded speech with shape {speech.shape}")
     noise, _ = load_audio(noise, sr)
     print(f"Loaded noise with shape {noise.shape}")
     speech, noise, noisy = mix_at_snr(speech, noise, snr)
+    print("Start denoising audio")
     enhanced = enhance(model, df, noisy)
+    print("Denoising finished")
     lim = torch.linspace(0.0, 1.0, int(sr * 0.15)).unsqueeze(0)
     lim = torch.cat((lim, torch.ones(1, enhanced.shape[1] - lim.shape[1])), dim=1)
     enhanced = enhanced * lim