Update app.py

app.py

@@ -3,13 +3,23 @@ import subprocess
 import openai
 import gradio as gr
 from gtts import gTTS
+import torch
+import librosa
+from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq
+
+# Load the processor and model
+processor = AutoProcessor.from_pretrained("lyimo/whisper-small-sw2")
+model = AutoModelForSpeechSeq2Seq.from_pretrained("lyimo/whisper-small-sw2")
 
 openai.api_key = os.getenv("OPENAI_API_KEY")
 
 def transcribe(audio):
-    with open(audio, "rb") as audio_file:
-        transcript = openai.Audio.transcribe("whisper-1", audio_file)
-    return transcript["text"]
+    audio_data, rate = librosa.load(audio, sr=16000)
+    inputs = processor(audio_data, sampling_rate=rate, return_tensors="pt")
+    with torch.no_grad():
+        outputs = model.generate(inputs["input_features"], max_length=512, num_beams=4, early_stopping=True)
+    transcription = [processor.decode(ids) for ids in outputs]
+    return transcription[0]
 
 def generate_response(transcribed_text):
     response = openai.ChatCompletion.create(