Order of inputs

demos/musicgen_app.py

@@ -244,8 +244,8 @@ def ui_full(launch_kwargs):
     with gr.Blocks() as interface:
         gr.Markdown(
             """
-            # MusicGen
-            Derived from [MusicGen](https://github.com/facebookresearch/audiocraft),
+            # Text-to-Music / Music-to-Music
+            MusicGen derived from [MusicGen](https://github.com/facebookresearch/audiocraft),
             presented at: ["Simple and Controllable Music Generation"](https://huggingface.co/papers/2306.05284).
             Modified for experiments in the re-scoring of *Metropolis* and *Wings*.
             """
@@ -260,9 +260,19 @@ def ui_full(launch_kwargs):
                         melody = gr.Audio(source="upload", type="numpy", label="File",
                                           interactive=True, elem_id="melody-input")
                 with gr.Row():
-                    submit = gr.Button("Generate", variant="primary")
+                    submit = gr.Button("Generate", variant = "primary")
                     # Adapted from https://github.com/rkfg/audiocraft/blob/long/app.py, MIT license.
-                    _ = gr.Button("Abort all", variant="stop").click(fn=interrupt, queue=False)
+                    _ = gr.Button("Abort all", variant="stop").click(fn=interrupt, queue = False)
+                with gr.Row():
+                    duration = gr.Slider(minimum = 1, maximum = 120, value = 10, label = "Duration", interactive = True)
+                with gr.Row():
+                    topk = gr.Number(label="Top-k", info = "Number of tokens shortlisted", value = 250, interactive = True)
+                    topp = gr.Number(label="Top-p", info = "Percent of tokens shortlisted", value = 0, interactive = True)
+                    temperature = gr.Number(label="Temperature", info = "lower=Always similar, higher=More creative", value = 1.0, interactive = True)
+                    cfg_coef = gr.Number(label="Classifier Free Guidance", value = 3.0, interactive=True)
+                with gr.Row():
+                    decoder = gr.Radio(["Default", "MultiBand_Diffusion"],
+                                       label="Decoder", value="Default", interactive=True)
                 with gr.Row():
                     model = gr.Radio(["facebook/musicgen-melody", "facebook/musicgen-medium", "facebook/musicgen-small",
                                       "facebook/musicgen-large", 
@@ -272,16 +282,6 @@ def ui_full(launch_kwargs):
                                       "facebook/musicgen-stereo-melody-large"],
                                      label="Model", value="facebook/musicgen-stereo-melody", interactive=True)
                     model_path = gr.Text(label="Model Path (custom models)")
-                with gr.Row():
-                    decoder = gr.Radio(["Default", "MultiBand_Diffusion"],
-                                       label="Decoder", value="Default", interactive=True)
-                with gr.Row():
-                    duration = gr.Slider(minimum=1, maximum=120, value=10, label="Duration", interactive=True)
-                with gr.Row():
-                    topk = gr.Number(label="Top-k", info="Number of tokens shortlisted", value=250, interactive=True)
-                    topp = gr.Number(label="Top-p", info="Percent of tokens shortlisted", value=0, interactive=True)
-                    temperature = gr.Number(label="Temperature", info="lower=Always similar, higher=More creative", value=1.0, interactive=True)
-                    cfg_coef = gr.Number(label="Classifier Free Guidance", value=3.0, interactive=True)
             with gr.Column():
                 output = gr.Video(label="Generated Music")
                 audio_output = gr.Audio(label="Generated Music (wav)", type='filepath')