Update app.py

app.py

@@ -2,7 +2,6 @@ import gradio as gr
 import whisper
 import os
 from pydub import AudioSegment
-from transformers import pipeline
 
 # Mapping of model names to Whisper model sizes
 MODELS = {
@@ -13,8 +12,8 @@ MODELS = {
     "Large (Most Accurate)": "large"
 }
 
-# Fine-tuned Sinhala model using Hugging Face pipeline
-SINHALA_PIPELINE = pipeline("automatic-speech-recognition", model="Subhaka/whisper-small-Sinhala-Fine_Tune")
+# Fine-tuned Sinhala model
+SINHALA_MODEL = "malakazzz/Subhaka-whisper-small-Sinhala-Fine_Tune"
 
 # Mapping of full language names to language codes
 LANGUAGE_NAME_TO_CODE = {
@@ -122,6 +121,14 @@ LANGUAGE_NAME_TO_CODE = {
 
 def transcribe_audio(audio_file, language="Auto Detect", model_size="Base (Faster)"):
     """Transcribe the audio file."""
+    # Load the appropriate model
+    if language == "Sinhala":
+        # Use the fine-tuned Sinhala model
+        model = gr.load(SINHALA_MODEL)
+    else:
+        # Use the selected Whisper model
+        model = whisper.load_model(MODELS[model_size])
+    
     # Convert audio to 16kHz mono for better compatibility with Whisper
     audio = AudioSegment.from_file(audio_file)
     audio = audio.set_frame_rate(16000).set_channels(1)
@@ -129,20 +136,13 @@ def transcribe_audio(audio_file, language="Auto Detect", model_size="Base (Faste
     audio.export(processed_audio_path, format="wav")
     
     # Transcribe the audio
-    if language == "Sinhala":
-        # Use the fine-tuned Sinhala model
-        result = SINHALA_PIPELINE(processed_audio_path)
-        detected_language = "si"
+    if language == "Auto Detect":
+        result = model.transcribe(processed_audio_path, fp16=False)  # Auto-detect language
+        detected_language = result.get("language", "unknown")
     else:
-        # Use the selected Whisper model
-        model = whisper.load_model(MODELS[model_size])
-        if language == "Auto Detect":
-            result = model.transcribe(processed_audio_path, fp16=False)  # Auto-detect language
-            detected_language = result.get("language", "unknown")
-        else:
-            language_code = LANGUAGE_NAME_TO_CODE.get(language, "en")  # Default to English if not found
-            result = model.transcribe(processed_audio_path, language=language_code, fp16=False)
-            detected_language = language_code
+        language_code = LANGUAGE_NAME_TO_CODE.get(language, "en")  # Default to English if not found
+        result = model.transcribe(processed_audio_path, language=language_code, fp16=False)
+        detected_language = language_code
     
     # Clean up processed audio file
     os.remove(processed_audio_path)
@@ -174,7 +174,7 @@ with gr.Blocks() as demo:
     # Update model dropdown based on language selection
     def update_model_dropdown(language):
         if language == "Sinhala":
-            return gr.Dropdown(interactive=False, value="Fine-Tuned Sinhala Model")
+            return gr.Dropdown(interactive=False, value="Base (Faster)")  # Set a valid value
         else:
             return gr.Dropdown(choices=list(MODELS.keys()), interactive=True, value="Base (Faster)")