Fix: Corrected shape handling for separated sources, extracting each source as a separate 1D array

2025-01-28 23:34:08 +05:30
parent 9e7c01c19e
commit fae60a8120
4 changed files with 9 additions and 6 deletions
@@ -27,9 +27,13 @@ def separate(audio, model_name='mpariente/ConvTasNet_WHAM!_sepclean'):
        with torch.no_grad():
            separated_sources = model(audio_tensor)
        
-        # Convert separated sources to NumPy arrays and remove batch dimension
-        separated_sources_np = [src.squeeze(0).cpu().numpy() for src in separated_sources]
+        # Remove batch dimension
+        separated_sources = separated_sources.squeeze(0)  # Shape: (num_sources, num_samples)
        
-        return separated_sources_np
+        # Split into list of sources
+        separated_sources_np = separated_sources.cpu().numpy()  # Convert to NumPy
+        separated_sources_list = [separated_sources_np[i, :] for i in range(separated_sources_np.shape[0])]
+        
+        return separated_sources_list
    except Exception as e:
        raise RuntimeError(f"Error during separation: {e}")