Update app.py

app.py

@@ -1,9 +1,39 @@
 import os
+import re
+import random
+from scipy.io.wavfile import write
+from scipy.io.wavfile import read
+import numpy as np
 import gradio as gr
+import yt_dlp
+import subprocess
 from pydub import AudioSegment
 from audio_separator.separator import Separator
 from lib.infer import infer_audio
 
+
+def download_audio(url):
+    ydl_opts = {
+        'format': 'bestaudio/best',
+        'outtmpl': 'ytdl/%(title)s.%(ext)s',
+        'postprocessors': [{
+            'key': 'FFmpegExtractAudio',
+            'preferredcodec': 'wav',
+            'preferredquality': '192',
+        }],
+    }
+
+    with yt_dlp.YoutubeDL(ydl_opts) as ydl:
+        info_dict = ydl.extract_info(url, download=True)
+        file_path = ydl.prepare_filename(info_dict).rsplit('.', 1)[0] + '.wav'
+        sample_rate, audio_data = read(file_path)
+        audio_array = np.asarray(audio_data, dtype=np.int16)
+
+        return sample_rate, audio_array
+
+
+
+
 # Define a function to handle the entire separation process
 def separate_audio(input_audio, output_dir, model_voc_inst, model_deecho, model_back_voc):
     # Create output directory if it doesn't exist
@@ -89,32 +119,6 @@ def process_audio(MODEL_NAME, SOUND_PATH, F0_CHANGE, F0_METHOD, MIN_PITCH, MAX_P
 with gr.Blocks(title="Hex RVC") as app:
     gr.Markdown("# Hex RVC")
     
-    with gr.Tab("Audio Separation"):
-        with gr.Row():
-            input_audio = gr.Audio(source="upload", type="filepath", label="Upload Audio File")
-            output_dir = gr.Textbox(value="/content/output", label="Output Directory")
-        
-        with gr.Row():
-            model_voc_inst = gr.Textbox(value='model_bs_roformer_ep_317_sdr_12.9755.ckpt', label="Vocal & Instrumental Model")
-            model_deecho = gr.Textbox(value='UVR-DeEcho-DeReverb.pth', label="DeEcho-DeReverb Model")
-            model_back_voc = gr.Textbox(value='mel_band_roformer_karaoke_aufr33_viperx_sdr_10.1956.ckpt', label="Backing Vocals Model")
-        
-        separate_button = gr.Button("Separate Audio")
-        
-        with gr.Row():
-            instrumental_out = gr.Audio(label="Instrumental")
-            vocals_out = gr.Audio(label="Vocals")
-            vocals_reverb_out = gr.Audio(label="Vocals (Reverb)")
-            vocals_no_reverb_out = gr.Audio(label="Vocals (No Reverb)")
-            lead_vocals_out = gr.Audio(label="Lead Vocals")
-            backing_vocals_out = gr.Audio(label="Backing Vocals")
-        
-        separate_button.click(
-            separate_audio,
-            inputs=[input_audio, output_dir, model_voc_inst, model_deecho, model_back_voc],
-            outputs=[instrumental_out, vocals_out, vocals_reverb_out, vocals_no_reverb_out, lead_vocals_out, backing_vocals_out]
-        )
-    
     with gr.Tab("Inference"):
         with gr.Row():
             MODEL_NAME = gr.Textbox(label="Model Name", placeholder="Enter model name")
@@ -160,5 +164,49 @@ with gr.Blocks(title="Hex RVC") as app:
             outputs=output_audio
         )
 
+    with gr.Tab("Audio Separation"):
+        with gr.Row():
+            input_audio = gr.Audio(source="upload", type="filepath", label="Upload Audio File")
+            output_dir = gr.Textbox(value="/content/output", label="Output Directory")
+
+            with gr.Accordion("Separation by Link", open = False):
+                with gr.Row():
+                    roformer_link = gr.Textbox(
+                    label = "Link",
+                    placeholder = "Paste the link here",
+                    interactive = True
+                )
+                with gr.Row():
+                   gr.Markdown("You can paste the link to the video/audio from many sites, check the complete list [here](https://github.com/yt-dlp/yt-dlp/blob/master/supportedsites.md)")
+                with gr.Row():
+                    roformer_download_button = gr.Button(
+                    "Download!",
+                    variant = "primary"
+                )
+
+            roformer_download_button.click(download_audio, [roformer_link], [input_audio])
+        
+        with gr.Row():
+            model_voc_inst = gr.Textbox(value='model_bs_roformer_ep_317_sdr_12.9755.ckpt', label="Vocal & Instrumental Model")
+            model_deecho = gr.Textbox(value='UVR-DeEcho-DeReverb.pth', label="DeEcho-DeReverb Model")
+            model_back_voc = gr.Textbox(value='mel_band_roformer_karaoke_aufr33_viperx_sdr_10.1956.ckpt', label="Backing Vocals Model")
+        
+        separate_button = gr.Button("Separate Audio")
+        
+        with gr.Row():
+            instrumental_out = gr.Audio(label="Instrumental")
+            vocals_out = gr.Audio(label="Vocals")
+            vocals_reverb_out = gr.Audio(label="Vocals (Reverb)")
+            vocals_no_reverb_out = gr.Audio(label="Vocals (No Reverb)")
+            lead_vocals_out = gr.Audio(label="Lead Vocals")
+            backing_vocals_out = gr.Audio(label="Backing Vocals")
+        
+        separate_button.click(
+            separate_audio,
+            inputs=[input_audio, output_dir, model_voc_inst, model_deecho, model_back_voc],
+            outputs=[instrumental_out, vocals_out, vocals_reverb_out, vocals_no_reverb_out, lead_vocals_out, backing_vocals_out]
+        )
+
+
 # Launch the Gradio app
 app.launch()