Add model documentation

README.md

@@ -70,7 +70,8 @@ def transcribe_audio(audio_path):
     input_features = processor(
         waveform.squeeze().numpy(),
         sampling_rate=16000,
-        return_tensors="pt"
+        return_tensors="pt",
+        language="uz"
     ).input_features.to(device)
 
     # Generate transcription