Update app.py

app.py

@@ -9,7 +9,7 @@ import soundfile as sf
 transcriber = pipeline("automatic-speech-recognition", model="facebook/s2t-small-librispeech-asr")
 
 # Initialize LLM pipeline
-generator = pipeline("text-generation", model="gpt2")
+generator = pipeline("text-generation", model="microsoft/Phi-3-mini-4k-instruct")
 
 # Initialize TTS tokenizer and model
 tokenizer = VitsTokenizer.from_pretrained("facebook/mms-tts-eng")
@@ -33,7 +33,7 @@ def transcribe_and_generate_audio(audio):
         outputs = model(**inputs)
     waveform = outputs.waveform[0]
     waveform_path = "output.wav"
-    sf.write(waveform_path, waveform.numpy(), 22050, format='wav')
+    sf.write(waveform_path, waveform.numpy(), 16000, format='wav')
 
     return waveform_path