Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -20,9 +20,10 @@ ROFORMER_MODELS = {
|
|
20 |
'Mel-Roformer-Karaoke-Aufr33-Viperx.ckpt': 'mel_band_roformer_karaoke_aufr33_viperx_sdr_10.1956.ckpt',
|
21 |
'Mel-Roformer-Denoise-Aufr33': 'denoise_mel_band_roformer_aufr33_sdr_27.9959.ckpt',
|
22 |
'Mel-Roformer-Denoise-Aufr33-Aggr': 'denoise_mel_band_roformer_aufr33_aggr_sdr_27.9768.ckpt',
|
23 |
-
'
|
24 |
-
'
|
25 |
-
'
|
|
|
26 |
}
|
27 |
MDX23C_MODELS = [
|
28 |
'MDX23C_D1581.ckpt',
|
@@ -109,7 +110,7 @@ def print_message(input_file, model_name):
|
|
109 |
"""Prints information about the audio separation process."""
|
110 |
base_name = os.path.splitext(os.path.basename(input_file))[0]
|
111 |
print("\n")
|
112 |
-
print("🎵
|
113 |
print("Input audio:", base_name)
|
114 |
print("Separation Model:", model_name)
|
115 |
print("Audio Separation Process...")
|
@@ -323,15 +324,15 @@ with gr.Blocks(
|
|
323 |
gr.HTML("<h1> 🎵 Audio-Separator 🎵 </h1>")
|
324 |
with gr.Accordion("General settings", open=False):
|
325 |
with gr.Group():
|
326 |
-
model_file_dir = gr.Textbox(value="/tmp/audio-separator-models/", label="Directory to cache model files", info="The directory where model files are stored.", placeholder="/tmp/audio-separator-models/"
|
327 |
with gr.Row():
|
328 |
-
output_dir = gr.Textbox(value="output", label="File output directory", info="The directory where output files will be saved.", placeholder="output"
|
329 |
output_format = gr.Dropdown(value="wav", choices=["wav", "flac", "mp3"], label="Output Format", info="The format of the output audio file.")
|
330 |
with gr.Row():
|
331 |
norm_threshold = gr.Slider(minimum=0.1, maximum=1, step=0.1, value=0.9, label="Normalization threshold", info="The threshold for audio normalization.")
|
332 |
amp_threshold = gr.Slider(minimum=0.1, maximum=1, step=0.1, value=0.6, label="Amplification threshold", info="The threshold for audio amplification.")
|
333 |
with gr.Row():
|
334 |
-
batch_size = gr.Slider(minimum=1, maximum=16, step=1, value=1, label="Batch Size", info="Larger consumes more RAM but may process slightly faster."
|
335 |
|
336 |
with gr.Tab("Roformer"):
|
337 |
with gr.Group():
|
@@ -520,4 +521,8 @@ with gr.Blocks(
|
|
520 |
outputs=[demucs_stem1, demucs_stem2, demucs_stem3, demucs_stem4, demucs_stem5, demucs_stem6],
|
521 |
)
|
522 |
|
523 |
-
|
|
|
|
|
|
|
|
|
|
20 |
'Mel-Roformer-Karaoke-Aufr33-Viperx.ckpt': 'mel_band_roformer_karaoke_aufr33_viperx_sdr_10.1956.ckpt',
|
21 |
'Mel-Roformer-Denoise-Aufr33': 'denoise_mel_band_roformer_aufr33_sdr_27.9959.ckpt',
|
22 |
'Mel-Roformer-Denoise-Aufr33-Aggr': 'denoise_mel_band_roformer_aufr33_aggr_sdr_27.9768.ckpt',
|
23 |
+
'MelBand Roformer Kim | Inst V1 by Unwa': 'melband_roformer_inst_v1.ckpt',
|
24 |
+
'MelBand Roformer Kim | Inst V2 by Unwa': 'melband_roformer_inst_v2.ckpt',
|
25 |
+
'MelBand Roformer Kim | InstVoc Duality V1 by Unwa': 'melband_roformer_instvoc_duality_v1.ckpt',
|
26 |
+
'MelBand Roformer Kim | InstVoc Duality V2 by Unwa': 'melband_roformer_instvox_duality_v2.ckpt',
|
27 |
}
|
28 |
MDX23C_MODELS = [
|
29 |
'MDX23C_D1581.ckpt',
|
|
|
110 |
"""Prints information about the audio separation process."""
|
111 |
base_name = os.path.splitext(os.path.basename(input_file))[0]
|
112 |
print("\n")
|
113 |
+
print("🎵 Audio-Separator 🎵")
|
114 |
print("Input audio:", base_name)
|
115 |
print("Separation Model:", model_name)
|
116 |
print("Audio Separation Process...")
|
|
|
324 |
gr.HTML("<h1> 🎵 Audio-Separator 🎵 </h1>")
|
325 |
with gr.Accordion("General settings", open=False):
|
326 |
with gr.Group():
|
327 |
+
model_file_dir = gr.Textbox(value="/tmp/audio-separator-models/", label="Directory to cache model files", info="The directory where model files are stored.", placeholder="/tmp/audio-separator-models/")
|
328 |
with gr.Row():
|
329 |
+
output_dir = gr.Textbox(value="output", label="File output directory", info="The directory where output files will be saved.", placeholder="output")
|
330 |
output_format = gr.Dropdown(value="wav", choices=["wav", "flac", "mp3"], label="Output Format", info="The format of the output audio file.")
|
331 |
with gr.Row():
|
332 |
norm_threshold = gr.Slider(minimum=0.1, maximum=1, step=0.1, value=0.9, label="Normalization threshold", info="The threshold for audio normalization.")
|
333 |
amp_threshold = gr.Slider(minimum=0.1, maximum=1, step=0.1, value=0.6, label="Amplification threshold", info="The threshold for audio amplification.")
|
334 |
with gr.Row():
|
335 |
+
batch_size = gr.Slider(minimum=1, maximum=16, step=1, value=1, label="Batch Size", info="Larger consumes more RAM but may process slightly faster.")
|
336 |
|
337 |
with gr.Tab("Roformer"):
|
338 |
with gr.Group():
|
|
|
521 |
outputs=[demucs_stem1, demucs_stem2, demucs_stem3, demucs_stem4, demucs_stem5, demucs_stem6],
|
522 |
)
|
523 |
|
524 |
+
def main():
|
525 |
+
app.launch(share=True)
|
526 |
+
|
527 |
+
if __name__ == "__main__":
|
528 |
+
main()
|