Updating Wiki tabs in AudioGen and MusicGen

app.py

@@ -959,11 +959,12 @@ def ui_full(launch_kwargs):
             Welcome to Soundscapes - TulipAI’s flagship Audio Storytelling Toolkit. Designed with modern content creators in mind, our AI-driven platform generates audio sound effects in just minutes tailored to your unique needs.
             """
         )
-        with gr.Tab("AudioGen"):
+        with gr.Tab("## AudioGen"):
             with gr.Tab("Generate Sound Effects"):
                 gr.Markdown(
                     """
                     ### AudioGen
+                    Check the "Wiki" to learn how to take the most out of TulipAI Soundscapes Sound Effects Generation Tool.
                     """
                 )
                 with gr.Row():
@@ -1052,7 +1053,7 @@ def ui_full(launch_kwargs):
                                 send_audio_a = gr.Button("Send to Input Audio")
                             seed_used_a = gr.Number(label='Seed used', value=-1, interactive=False)
                             download_a = gr.File(label="Generated Files", interactive=False)
-            with gr.Tab("Wiki"):
+            with gr.Tab("## Wiki"):
                         gr.Markdown(
                             """
                             - **[Generate (button)]:**  
@@ -1179,111 +1180,113 @@ def ui_full(launch_kwargs):
                             """
                         )
         with gr.Tab("MusicGen"):
-            gr.Markdown(
-                """
-                ### MusicGen
-                """
-            )
-            with gr.Row():
-                with gr.Column():
-                    with gr.Tab("Generation"):
-                        with gr.Accordion("Structure Prompts", open=False):
-                            with gr.Column():
-                                with gr.Row():
-                                    struc_prompts = gr.Checkbox(label="Enable", value=False, interactive=True, container=False)
-                                    bpm = gr.Number(label="BPM", value=120, interactive=True, scale=1, precision=0)
-                                    key = gr.Dropdown(["C", "C#", "D", "D#", "E", "F", "F#", "G", "G#", "A", "Bb", "B"], label="Key", value="C", interactive=True)
-                                    scale = gr.Dropdown(["Major", "Minor"], label="Scale", value="Major", interactive=True)
-                                with gr.Row():
-                                    global_prompt = gr.Text(label="Global Prompt", interactive=True, scale=3)
-                        with gr.Row():
-                            s = gr.Slider(1, max_textboxes, value=1, step=1, label="Prompts:", interactive=True, scale=2)
-                            #s_mode = gr.Radio(["segmentation", "batch"], value="segmentation", interactive=True, scale=1, label="Generation Mode")
-                        with gr.Column():
-                            textboxes = []
-                            prompts = []
-                            repeats = []
-                            calcs = []
+            gr.tab("## MusicGen"):
+                gr.Markdown(
+                    """
+                    ### MusicGen
+                    Check the "Wiki" to learn how to take the most out of TulipAI Soundscapes Music Generation Tool.
+                    """
+                )
+                with gr.Row():
+                    with gr.Column():
+                        with gr.Tab("Generation"):
+                            with gr.Accordion("Structure Prompts", open=False):
+                                with gr.Column():
+                                    with gr.Row():
+                                        struc_prompts = gr.Checkbox(label="Enable", value=False, interactive=True, container=False)
+                                        bpm = gr.Number(label="BPM", value=120, interactive=True, scale=1, precision=0)
+                                        key = gr.Dropdown(["C", "C#", "D", "D#", "E", "F", "F#", "G", "G#", "A", "Bb", "B"], label="Key", value="C", interactive=True)
+                                        scale = gr.Dropdown(["Major", "Minor"], label="Scale", value="Major", interactive=True)
+                                    with gr.Row():
+                                        global_prompt = gr.Text(label="Global Prompt", interactive=True, scale=3)
                             with gr.Row():
-                                text0 = gr.Text(label="Input Text", interactive=True, scale=4)
-                                prompts.append(text0)
-                                drag0 = gr.Number(label="Repeat", value=1, interactive=True, scale=1)
-                                repeats.append(drag0)
-                                calc0 = gr.Text(interactive=False, value="00:00 - 00:00", scale=1, label="Time")
-                                calcs.append(calc0)
-                            for i in range(max_textboxes):
-                                with gr.Row(visible=False) as t:
-                                    text = gr.Text(label="Input Text", interactive=True, scale=3)
-                                    repeat = gr.Number(label="Repeat", minimum=1, value=1, interactive=True, scale=1)
-                                    calc = gr.Text(interactive=False, value="00:00 - 00:00", scale=1, label="Time")
-                                textboxes.append(t)
-                                prompts.append(text)
-                                repeats.append(repeat)
-                                calcs.append(calc)
-                            to_calc = gr.Button("Calculate Timings", variant="secondary")
-                        with gr.Row():
-                            duration = gr.Slider(minimum=1, maximum=300, value=10, step=1, label="Duration", interactive=True)
-                        with gr.Row():
-                            overlap = gr.Slider(minimum=1, maximum=29, value=12, step=1, label="Overlap", interactive=True)
-                        with gr.Row():
-                            seed = gr.Number(label="Seed", value=-1, scale=4, precision=0, interactive=True)
-                            gr.Button('\U0001f3b2\ufe0f', scale=1).click(fn=lambda: -1, outputs=[seed], queue=False)
-                            reuse_seed = gr.Button('\u267b\ufe0f', scale=1)
-
-                    with gr.Tab("Audio"):
-                        with gr.Row():
-                            with gr.Column():
-                                input_type = gr.Radio(["file", "mic"], value="file", label="Input Type (optional)", interactive=True)
-                                mode = gr.Radio(["melody", "sample"], label="Input Audio Mode (optional)", value="sample", interactive=True)
-                                with gr.Row():
-                                    trim_start = gr.Number(label="Trim Start", value=0, interactive=True)
-                                    trim_end = gr.Number(label="Trim End", value=0, interactive=True)
-                            audio = gr.Audio(source="upload", type="numpy", label="Input Audio (optional)", interactive=True)
-
-                    with gr.Tab("Customization"):
-                        with gr.Row():
-                            with gr.Column():
-                                background = gr.ColorPicker(value="#0f0f0f", label="background color", interactive=True, scale=0)
-                                bar1 = gr.ColorPicker(value="#84cc16", label="bar color start", interactive=True, scale=0)
-                                bar2 = gr.ColorPicker(value="#10b981", label="bar color end", interactive=True, scale=0)
+                                s = gr.Slider(1, max_textboxes, value=1, step=1, label="Prompts:", interactive=True, scale=2)
+                                #s_mode = gr.Radio(["segmentation", "batch"], value="segmentation", interactive=True, scale=1, label="Generation Mode")
                             with gr.Column():
-                                image = gr.Image(label="Background Image", type="filepath", interactive=True, scale=4)
+                                textboxes = []
+                                prompts = []
+                                repeats = []
+                                calcs = []
                                 with gr.Row():
-                                    height = gr.Number(label="Height", value=512, interactive=True)
-                                    width = gr.Number(label="Width", value=768, interactive=True)
-
-                    with gr.Tab("Settings"):
-                        with gr.Row():
-                            channel = gr.Radio(["mono", "stereo", "stereo effect"], label="Output Audio Channels", value="stereo", interactive=True, scale=1)
-                            sr_select = gr.Dropdown(["11025", "16000", "22050", "24000", "32000", "44100", "48000"], label="Output Audio Sample Rate", value="48000", interactive=True)
-                        with gr.Row():
-                            model = gr.Radio(["melody", "small", "medium", "large", "custom"], label="Model", value="large", interactive=True, scale=1)
-                            with gr.Column():
-                                dropdown = gr.Dropdown(choices=get_available_models(), value=("No models found" if len(get_available_models()) < 1 else get_available_models()[0]), label='Custom Model (models folder)', elem_classes='slim-dropdown', interactive=True)
-                                ui.create_refresh_button(dropdown, lambda: None, lambda: {'choices': get_available_models()}, 'refresh-button')
-                                basemodel = gr.Radio(["small", "medium", "melody", "large"], label="Base Model", value="medium", interactive=True, scale=1)
-                        with gr.Row():
-                            decoder = gr.Radio(["Default", "MultiBand_Diffusion"], label="Decoder", value="Default", interactive=True)
-                        with gr.Row():
-                            topk = gr.Number(label="Top-k", value=250, interactive=True)
-                            topp = gr.Number(label="Top-p", value=0, interactive=True)
-                            temperature = gr.Number(label="Temperature", value=1.0, interactive=True)
-                            cfg_coef = gr.Number(label="Classifier Free Guidance", value=3.0, interactive=True)
-                    with gr.Row():
-                        submit = gr.Button("Generate", variant="primary")
-                        # Adapted from https://github.com/rkfg/audiocraft/blob/long/app.py, MIT license.
-                        _ = gr.Button("Interrupt").click(fn=interrupt, queue=False)
-                with gr.Column() as c:
-                    with gr.Tab("Output"):
-                        output = gr.Video(label="Generated Music", scale=0)
+                                    text0 = gr.Text(label="Input Text", interactive=True, scale=4)
+                                    prompts.append(text0)
+                                    drag0 = gr.Number(label="Repeat", value=1, interactive=True, scale=1)
+                                    repeats.append(drag0)
+                                    calc0 = gr.Text(interactive=False, value="00:00 - 00:00", scale=1, label="Time")
+                                    calcs.append(calc0)
+                                for i in range(max_textboxes):
+                                    with gr.Row(visible=False) as t:
+                                        text = gr.Text(label="Input Text", interactive=True, scale=3)
+                                        repeat = gr.Number(label="Repeat", minimum=1, value=1, interactive=True, scale=1)
+                                        calc = gr.Text(interactive=False, value="00:00 - 00:00", scale=1, label="Time")
+                                    textboxes.append(t)
+                                    prompts.append(text)
+                                    repeats.append(repeat)
+                                    calcs.append(calc)
+                                to_calc = gr.Button("Calculate Timings", variant="secondary")
+                            with gr.Row():
+                                duration = gr.Slider(minimum=1, maximum=300, value=10, step=1, label="Duration", interactive=True)
+                            with gr.Row():
+                                overlap = gr.Slider(minimum=1, maximum=29, value=12, step=1, label="Overlap", interactive=True)
+                            with gr.Row():
+                                seed = gr.Number(label="Seed", value=-1, scale=4, precision=0, interactive=True)
+                                gr.Button('\U0001f3b2\ufe0f', scale=1).click(fn=lambda: -1, outputs=[seed], queue=False)
+                                reuse_seed = gr.Button('\u267b\ufe0f', scale=1)
+    
+                        with gr.Tab("Audio"):
+                            with gr.Row():
+                                with gr.Column():
+                                    input_type = gr.Radio(["file", "mic"], value="file", label="Input Type (optional)", interactive=True)
+                                    mode = gr.Radio(["melody", "sample"], label="Input Audio Mode (optional)", value="sample", interactive=True)
+                                    with gr.Row():
+                                        trim_start = gr.Number(label="Trim Start", value=0, interactive=True)
+                                        trim_end = gr.Number(label="Trim End", value=0, interactive=True)
+                                audio = gr.Audio(source="upload", type="numpy", label="Input Audio (optional)", interactive=True)
+    
+                        with gr.Tab("Customization"):
+                            with gr.Row():
+                                with gr.Column():
+                                    background = gr.ColorPicker(value="#0f0f0f", label="background color", interactive=True, scale=0)
+                                    bar1 = gr.ColorPicker(value="#84cc16", label="bar color start", interactive=True, scale=0)
+                                    bar2 = gr.ColorPicker(value="#10b981", label="bar color end", interactive=True, scale=0)
+                                with gr.Column():
+                                    image = gr.Image(label="Background Image", type="filepath", interactive=True, scale=4)
+                                    with gr.Row():
+                                        height = gr.Number(label="Height", value=512, interactive=True)
+                                        width = gr.Number(label="Width", value=768, interactive=True)
+    
+                        with gr.Tab("Settings"):
+                            with gr.Row():
+                                channel = gr.Radio(["mono", "stereo", "stereo effect"], label="Output Audio Channels", value="stereo", interactive=True, scale=1)
+                                sr_select = gr.Dropdown(["11025", "16000", "22050", "24000", "32000", "44100", "48000"], label="Output Audio Sample Rate", value="48000", interactive=True)
+                            with gr.Row():
+                                model = gr.Radio(["melody", "small", "medium", "large", "custom"], label="Model", value="large", interactive=True, scale=1)
+                                with gr.Column():
+                                    dropdown = gr.Dropdown(choices=get_available_models(), value=("No models found" if len(get_available_models()) < 1 else get_available_models()[0]), label='Custom Model (models folder)', elem_classes='slim-dropdown', interactive=True)
+                                    ui.create_refresh_button(dropdown, lambda: None, lambda: {'choices': get_available_models()}, 'refresh-button')
+                                    basemodel = gr.Radio(["small", "medium", "melody", "large"], label="Base Model", value="medium", interactive=True, scale=1)
+                            with gr.Row():
+                                decoder = gr.Radio(["Default", "MultiBand_Diffusion"], label="Decoder", value="Default", interactive=True)
+                            with gr.Row():
+                                topk = gr.Number(label="Top-k", value=250, interactive=True)
+                                topp = gr.Number(label="Top-p", value=0, interactive=True)
+                                temperature = gr.Number(label="Temperature", value=1.0, interactive=True)
+                                cfg_coef = gr.Number(label="Classifier Free Guidance", value=3.0, interactive=True)
                         with gr.Row():
-                            audio_only = gr.Audio(type="numpy", label="Audio Only", interactive=False)
-                            backup_only = gr.Audio(type="numpy", label="Backup Audio", interactive=False, visible=False)
-                            send_audio = gr.Button("Send to Input Audio")
-                        seed_used = gr.Number(label='Seed used', value=-1, interactive=False)
-                        download = gr.File(label="Generated Files", interactive=False)
-                    with gr.Tab("Wiki"):
-                        gr.Markdown(
+                            submit = gr.Button("Generate", variant="primary")
+                            # Adapted from https://github.com/rkfg/audiocraft/blob/long/app.py, MIT license.
+                            _ = gr.Button("Interrupt").click(fn=interrupt, queue=False)
+                    with gr.Column() as c:
+                        with gr.Tab("Output"):
+                            output = gr.Video(label="Generated Music", scale=0)
+                            with gr.Row():
+                                audio_only = gr.Audio(type="numpy", label="Audio Only", interactive=False)
+                                backup_only = gr.Audio(type="numpy", label="Backup Audio", interactive=False, visible=False)
+                                send_audio = gr.Button("Send to Input Audio")
+                            seed_used = gr.Number(label='Seed used', value=-1, interactive=False)
+                            download = gr.File(label="Generated Files", interactive=False)
+            with gr.Tab("## Wiki"):
+                gr.Markdown(
                             """
                             - **[Generate (button)]:**  
                             Generates the music with the given settings and prompts.