Update dub.py

dub.py

@@ -3,26 +3,13 @@ import tempfile
 from gtts import gTTS
 from pydub import AudioSegment
 import moviepy.editor as mp
-from vosk import Model, KaldiRecognizer
-import json
-
-model = Model("vosk-model-es-0.10")
-
-def text_to_speech(text, lang='es'):
-    tts = gTTS(text=text, lang=lang, slow=False)
-    temp_file = tempfile.NamedTemporaryFile(delete=False, suffix=".mp3")
-    tts.save(temp_file.name)
-    return temp_file.name
-
-def play_audio(audio_file):
-    sound = AudioSegment.from_mp3(audio_file)
-    sound.export("temp.wav", format="wav")
-    os.system("start temp.wav")
+import speech_recognition as sr
 
 def transcribe_and_dub(video_file):
     temp_folder = tempfile.mkdtemp()
-    recognizer = KaldiRecognizer(Model("vosk-model-es-0.10"), 16000)
     
+    recognizer = sr.Recognizer()
+
     with mp.VideoFileClip(video_file) as video:
         audio_file = os.path.join(temp_folder, "audio.wav")
         video.audio.write_audiofile(audio_file)
@@ -32,18 +19,14 @@ def transcribe_and_dub(video_file):
         
         audio_file = os.path.join(temp_folder, "audio16000.wav")
 
-        with open(audio_file, 'rb') as f:
-            data = f.read(1024)
-            while data:
-                recognizer.AcceptWaveform(data)
-                data = f.read(1024)
-    
-    result = json.loads(recognizer.Result())
-    text = result.get("text", "")
-    
+        with sr.AudioFile(audio_file) as source:
+            audio = recognizer.record(source)
+
+    text = recognizer.recognize_google(audio, language='es')
+
     dubbed_audio_file = text_to_speech(text)
     dubbed_video_file = os.path.join(temp_folder, "dubbed_video.mp4")
-    
+
     with mp.VideoFileClip(video_file) as original_video:
         dubbed_audio = AudioSegment.from_mp3(dubbed_audio_file)
         dubbed_audio = dubbed_audio[:len(original_video.audio)]