Update app.py

app.py

@@ -26,9 +26,42 @@ instruction = """
 #]
 
 
+recognizer = speech_recognition.Recognizer()
+vg = VoiceGenerator()
+
+
+#     while True:
+#         # старт записи речи с последующим выводом распознанной речи
+#         # и удалением записанного в микрофон аудио
+#         voice_input = record_and_recognize_audio()
+#         os.remove("microphone-results.wav")
+#         print(voice_input)
+#         path_to_file = vg.generate(ask(voice_input))
+#         print(path_to_file)
+#         response = requests.post(
+#             "https://api.gooey.ai/v2/Lipsync/form/",
+#             headers={
+#                 "Authorization": "Bearer " + k,
+#             },
+#             files=files,
+#             data={"json": json.dumps(payload)},
+#         )
+#         assert response.ok, response.content
+#         #song = AudioSegment.from_wav(path_to_file)
+#         result = response.json()
+#         print(response.status_code, result["output"]["output_video"])
+#         #play(song)
+#         urllib.request.urlretrieve(result["output"]["output_video"], "C:\\Users\\user\\Desktop\\deepfake_sirius\\materials\\video.mp4")
+#         os.startfile("C:\\Users\\user\\Desktop\\deepfake_sirius\\materials\\video.mp4")
+#         break;
+
+
 class VoiceGenerator:
+    
     def __init__(self):
         self.model = Model(model_path=PATH_TO_MODEL)
+
+        
     def generate(self, text, file_name='output.wav'):
         synth = Synth(self.model)
         path = os.path.join(PATH_TO_OUTPUT, file_name)
@@ -36,7 +69,7 @@ class VoiceGenerator:
         return path
 
 
-def record_and_recognize_audio(file_path):
+def recognize_audio(file_path):
     with speech_recognition.AudioFile(file_path) as source:
         audio = recognizer.record(source)
     try:
@@ -48,7 +81,7 @@ def record_and_recognize_audio(file_path):
     return recognized_data
 
 
-def ask(request):
+def ask_gpt(request):
     result = requests.post(
         url='https://llm.api.cloud.yandex.net/llm/v1alpha/instruct',
         headers={
@@ -69,11 +102,16 @@ def ask(request):
 
 
 def result(audio):
-    sr, data = audio
-    print('sr:', sr, 'data:', data)
+    sample_rate, data = audio
+    #print('sample_rate:', sample_rate, 'data:', data)
     #return os.path.join(os.path.abspath(''), "video_sample.mp4")
-    write('voice_input.wav', sr, data)
-    return os.path.join(os.path.abspath(''), "voice_input.wav")
+    file_name = 'voice_input.wav'
+    file_path = os.path.join(PATH_TO_OUTPUT, file_name)
+    write(file_path, sample_rate, data)
+    text_from_audio = recognize_audio(file_path)
+    generated_audio = vg.generate(text_from_audio)
+    return generated_audio
+
 
 demo = gr.Interface(
     result,