Create app.py

app.py

@@ -0,0 +1,41 @@
+import gradio as gr
+from transformers import WhisperProcessor, WhisperForConditionalGeneration
+import torch
+
+# Load the fine-tuned Whisper model and processor
+model_name = "hackergeek98/tinyyyy_whisper"
+processor = WhisperProcessor.from_pretrained(model_name)
+model = WhisperForConditionalGeneration.from_pretrained(model_name)
+
+# Move model to GPU if available
+device = "cuda" if torch.cuda.is_available() else "cpu"
+model.to(device)
+
+# Define the ASR function
+def transcribe_audio(audio):
+    # Load audio file
+    sampling_rate, audio_data = audio
+
+    # Preprocess the audio
+    inputs = processor(audio_data, sampling_rate=sampling_rate, return_tensors="pt").input_features.to(device)
+
+    # Generate transcription
+    with torch.no_grad():
+        predicted_ids = model.generate(inputs)
+
+    # Decode the transcription
+    transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)[0]
+    return transcription
+
+# Create the Gradio interface
+interface = gr.Interface(
+    fn=transcribe_audio,  # Function to call
+    inputs=gr.Audio(source="upload", type="numpy"),  # Input: Upload audio file
+    outputs=gr.Textbox(label="Transcription"),  # Output: Display transcription
+    title="Whisper ASR: Tinyyyy Model",
+    description="Upload an audio file, and the fine-tuned Whisper model will transcribe it.",
+    examples=["example1.wav", "example2.wav"],  # Example audio files
+)
+
+# Launch the app
+interface.launch()