Update app.py

app.py

@@ -616,7 +616,7 @@ def preprocess_img(input_image_path, raw_image_path, orientation_state, session_
 
     return input_img_path, raw_image_path
 
-def infer_example(image_path, audio_path, text, num_steps, raw_image_path, session_id = None, progress=gr.Progress(track_tqdm=True),):
+def infer_example(image_path, audio_path, text, raw_image_path, session_id = None, progress=gr.Progress(track_tqdm=True),):
 
     if session_id is None:
         session_id = uuid.uuid4().hex
@@ -624,7 +624,7 @@ def infer_example(image_path, audio_path, text, num_steps, raw_image_path, sessi
     limited_audio_path = preprocess_audio_first_5s_librosa(audio_path, True, session_id)
 
     image_path, _ = preprocess_img(image_path, image_path, [[720, 400]], session_id)
-    result = infer(image_path, limited_audio_path, text, [[720, 400]], num_steps, session_id, progress)
+    result = infer(image_path, limited_audio_path, text, [[720, 400]], 8, session_id, progress)
 
     return result
 
@@ -885,7 +885,6 @@ with gr.Blocks(css=css) as demo:
                             "examples/images/female-001.png",
                             "examples/audios/script.wav",
                             ADAPTIVE_PROMPT_TEMPLATES[1],
-                            8,
                             ''
                         ],
                         
@@ -894,7 +893,6 @@ with gr.Blocks(css=css) as demo:
                             "examples/images/male-001.png",
                             "examples/audios/denial.wav",
                             ADAPTIVE_PROMPT_TEMPLATES[1],
-                            8,
                             ''
                         ],
 
@@ -902,7 +900,6 @@ with gr.Blocks(css=css) as demo:
                             "examples/images/female-003.png",
                             "examples/audios/matcha.wav",
                             ADAPTIVE_PROMPT_TEMPLATES[1],
-                            8,
                             ''
                         ],
 
@@ -910,13 +907,12 @@ with gr.Blocks(css=css) as demo:
                             "examples/images/female-007.png",
                             "examples/audios/listen.wav",
                             ADAPTIVE_PROMPT_TEMPLATES[1],
-                            8,
                             ''
                         ],
 
                     ],
                     label="Cached Examples",
-                    inputs=[image_input, audio_input, text_input, num_steps, raw_img_text],
+                    inputs=[image_input, audio_input, text_input, raw_img_text],
                     outputs=[output_video],
                     fn=infer_example,
                     cache_examples=True