Handle audio_path being None

app.py

@@ -23,42 +23,45 @@ def denoise(model_name: str, audio_path: str) -> str:
     if torch.cuda.is_available():
         model = model.cuda()
 
-    stream_reader = torchaudio.io.StreamReader(audio_path)
-    stream_reader.add_basic_audio_stream(
-        frames_per_chunk=model.config.max_length,
-        sample_rate=model.config.sample_rate,
-        num_channels=1,
-    )
+    if audio_path:
+        stream_reader = torchaudio.io.StreamReader(audio_path)
+        stream_reader.add_basic_audio_stream(
+            frames_per_chunk=model.config.max_length,
+            sample_rate=model.config.sample_rate,
+            num_channels=1,
+        )
 
-    stream_writer = torchaudio.io.StreamWriter("denoised.wav")
-    stream_writer.add_audio_stream(sample_rate=model.config.sample_rate, num_channels=1)
+        stream_writer = torchaudio.io.StreamWriter("denoised.wav")
+        stream_writer.add_audio_stream(
+            sample_rate=model.config.sample_rate, num_channels=1
+        )
 
-    chunk_size = model.config.max_length
+        chunk_size = model.config.max_length
 
-    with stream_writer.open():
-        for (audio_chunk,) in tqdm(stream_reader.stream()):
-            if audio_chunk is None:
-                break
+        with stream_writer.open():
+            for (audio_chunk,) in tqdm(stream_reader.stream()):
+                if audio_chunk is None:
+                    break
 
-            audio_chunk = audio_chunk.permute(1, 0)
-            original_chunk_size = audio_chunk.size(-1)
+                audio_chunk = audio_chunk.permute(1, 0)
+                original_chunk_size = audio_chunk.size(-1)
 
-            if audio_chunk.size(-1) < chunk_size:
-                padding = chunk_size - audio_chunk.size(-1)
-                audio_chunk = torch.nn.functional.pad(audio_chunk, (0, padding))
+                if audio_chunk.size(-1) < chunk_size:
+                    padding = chunk_size - audio_chunk.size(-1)
+                    audio_chunk = torch.nn.functional.pad(audio_chunk, (0, padding))
 
-            if torch.cuda.is_available():
-                audio_chunk = audio_chunk.cuda()
+                if torch.cuda.is_available():
+                    audio_chunk = audio_chunk.cuda()
 
-            with torch.no_grad():
-                denoised_chunk = model(audio_chunk[None]).audio
-                denoised_chunk = denoised_chunk[:, :, :original_chunk_size]
+                with torch.no_grad():
+                    denoised_chunk = model(audio_chunk[None]).audio
+                    denoised_chunk = denoised_chunk[..., :original_chunk_size]
 
-            stream_writer.write_audio_chunk(
-                0, denoised_chunk.squeeze(0).permute(1, 0).cpu()
-            )
+                stream_writer.write_audio_chunk(
+                    0, denoised_chunk.squeeze(0).permute(1, 0).cpu()
+                )
 
-    return "denoised.wav"
+        return "denoised.wav"
 
 
 iface = gr.Interface(