add audios files

app.py

@@ -102,10 +102,14 @@ def enhance_speech(audio_array, sampling_rate, solver, nfe, tau, denoise_before_
     lambd = 0.9 if denoise_before_enhancement else 0.1
 
     denoised_audio, new_sr1 = denoise(audio_array, sampling_rate, device)
-    enhanced_audio, new_sr2 = enhance(audio_array, sampling_rate, device, nfe=nfe, solver=solver, lambd=lambd, tau=tau)
+    # enhanced_audio, new_sr2 = enhance(audio_array, sampling_rate, device, nfe=nfe, solver=solver, lambd=lambd, tau=tau)
 
     # Convert to numpy and return
-    return (new_sr1, denoised_audio.cpu().numpy()), (new_sr2, enhanced_audio.cpu().numpy())
+    return (
+        (new_sr1, denoised_audio.cpu().numpy()),
+        None
+        # (new_sr2, enhanced_audio.cpu().numpy())
+    )
 
 
 def convert_to_int16(audio_array):
@@ -138,21 +142,21 @@ def _fn(
     else:
         audio_array, sampling_rate = text_to_speech(bambara_text)
 
-    # # Step 3: Enhance the audio
-    # denoised_audio, enhanced_audio = enhance_speech(
-    #     audio_array,
-    #     sampling_rate,
-    #     solver,
-    #     nfe,
-    #     prior_temp,
-    #     denoise_before_enhancement
-    # )
+    # Step 3: Enhance the audio
+    denoised_audio, _ = enhance_speech(
+        audio_array,
+        sampling_rate,
+        solver,
+        nfe,
+        prior_temp,
+        denoise_before_enhancement
+    )
 
     # Return all outputs
     return (
         bambara_text,
         (sampling_rate, audio_array.numpy()),
-        # (denoised_audio[0], convert_to_int16(denoised_audio[1])),
+        denoised_audio,
         # (enhanced_audio[0], convert_to_int16(enhanced_audio[1]))
     )