Update app.py

app.py

@@ -34,8 +34,6 @@ def one_shot(image,input_text,gender):
           sound = AudioSegment.from_file(f.name, format="mp3")
           sound.export("/content/audio.wav", format="wav")
      waveform, sample_rate = torchaudio.load("/content/audio.wav")
-
-     
      torchaudio.save("/content/audio.wav", waveform, sample_rate, encoding="PCM_S", bits_per_sample=16)
      image = Image.open(image_in)
      image = pad_image(image)
@@ -47,7 +45,7 @@ def one_shot(image,input_text,gender):
     
      os.system(f"cd /content/one-shot-talking-face && python3 -B test_script.py --img_path /content/image_pre.png --audio_path /content/audio.wav --phoneme_path /content/test.json --save_dir /content/train")
      
-     return "/content/train/image_audio.mp4"
+     return "/content/image_pre.png"
      
        
    elif gender == 'Male' or gender == 'male':
@@ -90,7 +88,7 @@ def run():
           # audio_in = gr.Audio(show_label=False, type='filepath')
           input_text=gr.Textbox(lines=3, value="Hello How are you?", label="Input Text")
           gender = gr.Radio(["Female","Male"],value="Female",label="Gender")
-          video_out = gr.Video(label="output")
+          video_out = gr.Image(label="output")
           # video_out = gr.Video(show_label=False)
         with gr.Row().style(equal_height=True):
           btn = gr.Button("Generate")