New params for Roformers and MDX23C (#5)

- New params for Roformers and MDX23C (9865bea4ae88349cd0156edcbf30f5fb0fb2dfa8)


Co-authored-by: Eddycrack 864 <Eddycrack864@users.noreply.huggingface.co>

app.py

@@ -122,7 +122,7 @@ demucs_overlap_values = [
 ]
 
 @spaces.GPU(duration=300)
-def roformer_separator(roformer_audio, roformer_model, roformer_output_format, roformer_overlap):
+def roformer_separator(roformer_audio, roformer_model, roformer_output_format, roformer_overlap, roformer_segment_size):
   files_list = []
   files_list.clear()
   directory = "./outputs"
@@ -131,7 +131,7 @@ def roformer_separator(roformer_audio, roformer_model, roformer_output_format, r
   os.makedirs("outputs", exist_ok=True)
   write(f'{random_id}.wav', roformer_audio[0], roformer_audio[1])
   full_roformer_model = roformer_models[roformer_model]
-  prompt = f"audio-separator {random_id}.wav --model_filename {full_roformer_model} --output_dir=./outputs --output_format={roformer_output_format} --normalization=0.9 --mdxc_overlap={roformer_overlap}"
+  prompt = f"audio-separator {random_id}.wav --model_filename {full_roformer_model} --output_dir=./outputs --output_format={roformer_output_format} --normalization=0.9 --mdxc_overlap={roformer_overlap} --mdxc_segment_size={roformer_segment_size}"
   os.system(prompt)
 
   for file in os.listdir(directory):
@@ -144,7 +144,7 @@ def roformer_separator(roformer_audio, roformer_model, roformer_output_format, r
   return stem1_file, stem2_file
 
 @spaces.GPU(duration=300)
-def mdxc_separator(mdx23c_audio, mdx23c_model, mdx23c_output_format, mdx23c_segment_size, mdx23c_overlap):
+def mdxc_separator(mdx23c_audio, mdx23c_model, mdx23c_output_format, mdx23c_segment_size, mdx23c_overlap, mdx23c_denoise):
   files_list = []
   files_list.clear()
   directory = "./outputs"
@@ -153,6 +153,10 @@ def mdxc_separator(mdx23c_audio, mdx23c_model, mdx23c_output_format, mdx23c_segm
   os.makedirs("outputs", exist_ok=True)
   write(f'{random_id}.wav', mdx23c_audio[0], mdx23c_audio[1])
   prompt = f"audio-separator {random_id}.wav --model_filename {mdx23c_model} --output_dir=./outputs --output_format={mdx23c_output_format} --normalization=0.9 --mdxc_segment_size={mdx23c_segment_size} --mdxc_overlap={mdx23c_overlap}"
+  
+  if mdx23c_denoise:
+    prompt += " --mdx_enable_denoise"
+  
   os.system(prompt)
 
   for file in os.listdir(directory):
@@ -267,6 +271,15 @@ with gr.Blocks(theme="NoCrypt/miku@1.2.2", title="🎵 UVR5 UI 🎵") as app:
                     value = 4,
                     interactive = True
                 )
+                roformer_segment_size = gr.Slider(
+                    minimum = 32,
+                    maximum = 4000,
+                    step = 32,
+                    label = "Segment Size",
+                    info = "Larger consumes more resources, but may give better results.",
+                    value = 256,
+                    interactive = True
+                )
             with gr.Row():
                 roformer_audio = gr.Audio(
                     label = "Input Audio",
@@ -289,7 +302,7 @@ with gr.Blocks(theme="NoCrypt/miku@1.2.2", title="🎵 UVR5 UI 🎵") as app:
                     type = "filepath"
                 )
 
-            roformer_button.click(roformer_separator, [roformer_audio, roformer_model, roformer_output_format, roformer_overlap], [roformer_stem1, roformer_stem2])
+            roformer_button.click(roformer_separator, [roformer_audio, roformer_model, roformer_output_format, roformer_overlap, roformer_segment_size], [roformer_stem1, roformer_stem2])
         
         with gr.TabItem("MDX23C"):
             with gr.Row():
@@ -322,6 +335,12 @@ with gr.Blocks(theme="NoCrypt/miku@1.2.2", title="🎵 UVR5 UI 🎵") as app:
                     value = 8,
                     interactive = True
                 )
+                mdx23c_denoise = gr.Checkbox(
+                    label = "Denoise",
+                    info = "Enable denoising during separation.",
+                    value = False,
+                    interactive = True
+                )
             with gr.Row():
                 mdx23c_audio = gr.Audio(
                     label = "Input Audio",
@@ -344,7 +363,7 @@ with gr.Blocks(theme="NoCrypt/miku@1.2.2", title="🎵 UVR5 UI 🎵") as app:
                     type = "filepath"
                 )
 
-            mdx23c_button.click(mdxc_separator, [mdx23c_audio, mdx23c_model, mdx23c_output_format, mdx23c_segment_size, mdx23c_overlap], [mdx23c_stem1, mdx23c_stem2])
+            mdx23c_button.click(mdxc_separator, [mdx23c_audio, mdx23c_model, mdx23c_output_format, mdx23c_segment_size, mdx23c_overlap, mdx23c_denoise], [mdx23c_stem1, mdx23c_stem2])
         
         with gr.TabItem("MDX-NET"):
             with gr.Row():
@@ -544,7 +563,7 @@ with gr.Blocks(theme="NoCrypt/miku@1.2.2", title="🎵 UVR5 UI 🎵") as app:
               * Special thanks to [Ilaria](https://github.com/TheStingerX) for hosting this space and help.
               * Thanks to [Mikus](https://github.com/cappuch) for the help with the code.
               * Thanks to [Nick088](https://huggingface.co/Nick088) for the help to fix roformers.
-              * Improvements by [Blane187](https://github.com/Blane187).
+              * Improvements by [Blane187](https://huggingface.co/Blane187).
 
               You can donate to the original UVR5 project here:
               [!["Buy Me A Coffee"](https://www.buymeacoffee.com/assets/img/custom_images/orange_img.png)](https://www.buymeacoffee.com/uvr5)