Update app.py

app.py

@@ -124,62 +124,61 @@ with gr.Blocks() as demo:
         Generate and mix radio promos effortlessly with AI tools!
     """)
 
-    with gr.Row():
-        user_prompt = gr.Textbox(label="Promo Idea", placeholder="E.g., A 30-second promo for a morning show.")
-        llama_model_id = gr.Textbox(label="Llama Model ID", value="meta-llama/Meta-Llama-3-8B-Instruct")
-        duration = gr.Slider(label="Duration (seconds)", minimum=15, maximum=60, step=15, value=30)
-        audio_length = gr.Slider(label="Music Length (tokens)", minimum=128, maximum=1024, step=64, value=512)
-        speaker = gr.Textbox(label="Voice Style (optional)", placeholder="E.g., male, female, or neutral.")
-        ducking = gr.Checkbox(label="Enable Ducking", value=True)
-
-    generate_script_button = gr.Button("Generate Script")
-    script_output = gr.Textbox(label="Generated Script")
-    music_suggestion_output = gr.Textbox(label="Music Suggestion")
-    
-    generate_voice_button = gr.Button("Generate Voice")
-    voice_output = gr.Audio(label="Generated Voice", type="filepath")
-    
-    generate_music_button = gr.Button("Generate Music")
-    music_output = gr.Audio(label="Generated Music", type="filepath")
-    
-    blend_button = gr.Button("Blend Audio")
-    final_output = gr.Audio(label="Final Promo Audio", type="filepath")
-
-    def step_generate_script(user_prompt, model_id, duration):
-        return generate_script(user_prompt, model_id, hf_token, duration)
-
-    def step_generate_voice(script, speaker):
-        return generate_voice(script, speaker)
-
-    def step_generate_music(music_suggestion, audio_length):
-        return generate_music(music_suggestion, audio_length)
-
-    def step_blend_audio(voice_path, music_path, ducking):
-        return blend_audio(voice_path, music_path, ducking)
-
-    generate_script_button.click(
-        fn=lambda user_prompt, model_id, duration: generate_script(user_prompt, model_id, hf_token, duration),
-        inputs=[user_prompt, llama_model_id, duration],
-        outputs=[script_output, music_suggestion_output],
-    )
-
-    generate_voice_button.click(
-        fn=step_generate_voice,
-        inputs=[script_output, speaker],
-        outputs=[voice_output],
-    )
-
-    generate_music_button.click(
-        fn=step_generate_music,
-        inputs=[music_suggestion_output, audio_length],
-        outputs=[music_output],
-    )
-
-    blend_button.click(
-        fn=step_blend_audio,
-        inputs=[voice_output, music_output, ducking],
-        outputs=[final_output],
-    )
+    with gr.Tabs():
+        with gr.Tab("Step 1: Generate Script"):
+            with gr.Row():
+                user_prompt = gr.Textbox(label="Promo Idea", placeholder="E.g., A 30-second promo for a morning show.")
+                llama_model_id = gr.Textbox(label="Llama Model ID", value="meta-llama/Meta-Llama-3-8B-Instruct")
+                duration = gr.Slider(label="Duration (seconds)", minimum=15, maximum=60, step=15, value=30)
+
+            generate_script_button = gr.Button("Generate Script")
+            script_output = gr.Textbox(label="Generated Script")
+            music_suggestion_output = gr.Textbox(label="Music Suggestion")
+
+            generate_script_button.click(
+                fn=lambda user_prompt, model_id, duration: generate_script(user_prompt, model_id, hf_token, duration),
+                inputs=[user_prompt, llama_model_id, duration],
+                outputs=[script_output, music_suggestion_output],
+            )
+
+        with gr.Tab("Step 2: Generate Voice"):
+            with gr.Row():
+                speaker = gr.Textbox(label="Voice Style (optional)", placeholder="E.g., male, female, or neutral.")
+
+            generate_voice_button = gr.Button("Generate Voice")
+            voice_output = gr.Audio(label="Generated Voice", type="filepath")
+
+            generate_voice_button.click(
+                fn=lambda script, speaker: generate_voice(script, speaker),
+                inputs=[script_output, speaker],
+                outputs=[voice_output],
+            )
+
+        with gr.Tab("Step 3: Generate Music"):
+            with gr.Row():
+                audio_length = gr.Slider(label="Music Length (tokens)", minimum=128, maximum=1024, step=64, value=512)
+
+            generate_music_button = gr.Button("Generate Music")
+            music_output = gr.Audio(label="Generated Music", type="filepath")
+
+            generate_music_button.click(
+                fn=lambda music_suggestion, audio_length: generate_music(music_suggestion, audio_length),
+                inputs=[music_suggestion_output, audio_length],
+                outputs=[music_output],
+            )
+
+        with gr.Tab("Step 4: Blend Audio"):
+            with gr.Row():
+                ducking = gr.Checkbox(label="Enable Ducking", value=True)
+
+            blend_button = gr.Button("Blend Audio")
+            final_output = gr.Audio(label="Final Promo Audio", type="filepath")
+
+            blend_button.click(
+                fn=lambda voice_path, music_path, ducking: blend_audio(voice_path, music_path, ducking),
+                inputs=[voice_output, music_output, ducking],
+                outputs=[final_output],
+            )
 
     gr.Markdown("""
         <hr>
@@ -189,4 +188,3 @@ with gr.Blocks() as demo:
     """)
 
 demo.launch(debug=True)
-