Update app.py

app.py

@@ -24,21 +24,21 @@ def load_model(version):
     return MusicGen.get_pretrained(version)
 
 
-def predict(model, uploaded_image, melody, duration, topk, topp, temperature, cfg_coef):
+def predict(uploaded_image, melody, duration):
     text = img_to_text(uploaded_image, 'best', 4, fn_index=1)[0]
     global MODEL
     topk = int(topk)
-    if MODEL is None or MODEL.name != model:
-        MODEL = load_model(model)
+    if MODEL is None or MODEL.name != "melody":
+        MODEL = load_model("melody")
 
     if duration > MODEL.lm.cfg.dataset.segment_duration:
         raise gr.Error("MusicGen currently supports durations of up to 30 seconds!")
     MODEL.set_generation_params(
         use_sampling=True,
-        top_k=topk,
-        top_p=topp,
-        temperature=temperature,
-        cfg_coef=cfg_coef,
+        top_k=250,
+        top_p=0,
+        temperature=1.0,
+        cfg_coef=3.0,
         duration=duration,
     )
 
@@ -84,18 +84,18 @@ with gr.Blocks() as demo:
                 melody = gr.Audio(source="upload", type="numpy", label="Melody Condition (optional)", interactive=True)
             with gr.Row():
                 submit = gr.Button("Submit")
-            with gr.Row():
-                model = gr.Radio(["melody", "medium", "small", "large"], label="Model", value="melody", interactive=True)
+            #with gr.Row():
+            #   model = gr.Radio(["melody", "medium", "small", "large"], label="Model", value="melody", interactive=True)
             with gr.Row():
                 duration = gr.Slider(minimum=1, maximum=30, value=10, label="Duration", 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():
+            #   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.Column():
             output = gr.Audio(label="Generated Music")
-    submit.click(predict, inputs=[model, uploaded_image, melody, duration, topk, topp, temperature, cfg_coef], outputs=[output])
+    submit.click(predict, inputs=[uploaded_image, melody, duration, topk, topp, temperature, cfg_coef], outputs=[output])
     
     gr.Markdown(
         """
@@ -104,7 +104,7 @@ with gr.Blocks() as demo:
         The model will generate a short music extract based on the image you provided.
         You can generate up to 30 seconds of audio.
 
-        We present 4 model variations:
+        This demo is set to use only the Melody model
         1. Melody -- a music generation model capable of generating music condition on text and melody inputs. **Note**, you can also use text only.
         2. Small -- a 300M transformer decoder conditioned on text only.
         3. Medium -- a 1.5B transformer decoder conditioned on text only.