Update app.py

app.py

@@ -102,7 +102,7 @@ recognizer = speech_recognition.Recognizer()
 vg = VoiceGenerator()
 
 
-def get_video(pathToWav, pathToImage):
+def get_video(pathToWav, pathToImage, pathToResult):
     files = [
        ("input_face", open(pathToImage, "rb")),
        ("input_audio", open(pathToWav, "rb")),
@@ -117,6 +117,9 @@ def get_video(pathToWav, pathToImage):
         data={"json": json.dumps(payload)}
     )
     assert response.ok, response.content
+    result = response.json()
+    #print(response.status_code, result["output"]["output_video"])
+    urllib.request.urlretrieve(result["output"]["output_video"], pathToResult)
 
 
 def result(audio):
@@ -128,15 +131,18 @@ def result(audio):
     write(file_path, sample_rate, data)
     text_from_audio = recognize_audio(file_path)
     generated_audio = vg.generate(ask_gpt(text_from_audio))
-    referenceName = ''
-    pathToReference = os.path.join(os.getcwd(), os.path.join("content", referenceName))
-    return generated_audio
+    referenceName = 'reference.jpg'
+    resultName = 'video.mp4'
+    pathToReference = os.path.join(PATH_TO_OUTPUT, referenceName)
+    pathToResult = os.path.join(PATH_TO_OUTPUT, resultName)
+    get_video(generated_audio, pathToReference, pathToResult)
+    return pathToResult
 
 
 demo = gr.Interface(
     result,
     gr.Audio(sources=["microphone"]),
-    "audio", #playable_video
+    "playable_video", #playable_video
 )
 
 demo.launch()