Create app.py

app.py

@@ -0,0 +1,101 @@
+import gradio as gr
+import os, sys
+from colorama import Fore
+
+now_dir = os.getcwd()
+sys.path.append(now_dir)
+
+
+# Function to detect the .pth and .index files
+def detect_files(model_name):
+    model_dir = f"{now_dir}/assets/weights/{model_name}"
+    index_dir = f"{now_dir}/logs/{model_name}"
+
+    # Detect .pth file
+    model_pth_file = None
+    for file in os.listdir(model_dir):
+        if file.endswith(".pth"):
+            model_pth_file = os.path.join(model_dir, file)
+            break
+
+    # Detect .index file
+    index_file = None
+    for file in os.listdir(index_dir):
+        if file.endswith(".index"):
+            index_file = os.path.join(index_dir, file)
+            break
+
+    if model_pth_file and index_file:
+        return f"Model .pth file: {model_pth_file}\nIndex file: {index_file}"
+    else:
+        return "Model .pth or index file not found."
+
+# Function to process the audio using the detected files
+def process_audio(model_name, pitch, input_path, f0_method, save_as, index_rate, volume_normalization, consonant_protection):
+    model_dir = f"{now_dir}/assets/weights/{model_name}"
+    index_dir = f"{now_dir}/logs/{model_name}"
+
+    # Detect files
+    model_pth_file = None
+    index_file = None
+    for file in os.listdir(model_dir):
+        if file.endswith(".pth"):
+            model_pth_file = os.path.join(model_dir, file)
+            break
+
+    for file in os.listdir(index_dir):
+        if file.endswith(".index"):
+            index_file = os.path.join(index_dir, file)
+            break
+
+    if not model_pth_file or not index_file:
+        return "Model .pth or index file not found.", None
+
+    if not os.path.exists(input_path):
+        return f"{input_path} was not found in your RVC folder.", None
+
+    # Set environment variables for paths
+    os.environ['index_root'] = os.path.dirname(index_file)
+    index_path = os.path.basename(index_file)
+    os.environ['weight_root'] = os.path.dirname(model_pth_file)
+
+    # Remove any previous output
+    if os.path.exists(save_as):
+        os.remove(save_as)
+
+    # Execute the CLI command
+    os.system(f"python {now_dir}/tools/infer_cli.py --f0up_key {pitch} --input_path {input_path} --index_path {index_path} --f0method {f0_method} --opt_path {save_as} --model_name {model_name} --index_rate {index_rate} --device 'cuda:0' --is_half True --filter_radius 3 --resample_sr 0 --rms_mix_rate {volume_normalization} --protect {consonant_protection}")
+
+    if os.path.exists(save_as):
+        return "Processing complete. Here is your output audio:", save_as
+    else:
+        return "Error in processing audio.", None
+
+# Gradio interface
+with gr.Blocks() as demo:
+    gr.Markdown("# 🔊 **LISTEN TO YOUR MODEL**")
+
+    model_name = gr.Textbox(label="Model Name", value="Ren")
+    pitch = gr.Slider(minimum=-12, maximum=12, step=1, label="Pitch", value=0)
+    input_path = gr.Dropdown(label="",choices=show_available('audios'),value='',interactive=True)
+    f0_method = gr.Radio(choices=["rmvpe", "pm", "crepe"], label="F0 Method", value="rmvpe")
+    save_as = gr.Textbox(label="Save As", value="/content/RVC/audios/cli_output.wav")
+    index_rate = gr.Slider(minimum=0, maximum=1, step=0.01, label="Index Rate", value=0.5)
+    volume_normalization = gr.Slider(minimum=0, maximum=1, step=0.01, label="Volume Normalization", value=0)
+    consonant_protection = gr.Slider(minimum=0, maximum=1, step=0.01, label="Consonant Protection", value=0.5)
+
+    output_text = gr.Textbox(label="Output")
+    output_audio = gr.Audio(label="Processed Audio")
+
+    # Button to detect files
+    detect_btn = gr.Button("Detect Files")
+    detect_btn.click(fn=detect_files, inputs=[model_name], outputs=output_text)
+
+    # Button to process the audio and return audio output
+    submit_btn = gr.Button("Submit")
+    submit_btn.click(fn=process_audio, 
+                     inputs=[model_name, pitch, input_path, f0_method, save_as, index_rate, volume_normalization, consonant_protection], 
+                     outputs=[output_text, output_audio])
+
+# Launch the app
+demo.launch()