add gradio app

app.py

@@ -0,0 +1,48 @@
+import gradio as gr
+from speechbrain.inference.ASR import EncoderASR
+from pydub import AudioSegment
+import os
+
+# Load the ASR model
+asr_model = EncoderASR.from_hparams(
+    source="YosefA/wave2vec2_amharic_stt", 
+    savedir="pretrained_models/asr-wav2vec2-amharic"
+)
+
+# Directory to store converted audio files
+os.makedirs("temp_audio", exist_ok=True)
+
+def transcribe_audio(audio_file):
+    """
+    Converts the uploaded audio to .wav format, runs transcription, and returns the result.
+    """
+    # Save the uploaded audio to a temporary location
+    temp_audio_path = "temp_audio/input_audio.wav"
+    
+    # Convert audio to .wav format
+    sound = AudioSegment.from_file(audio_file)
+    sound.export(temp_audio_path, format="wav")
+    
+    # Transcribe the audio
+    transcription = asr_model.transcribe_file(temp_audio_path)
+    
+    # Clean up temporary files (optional)
+    os.remove(temp_audio_path)
+    
+    return transcription
+
+# Define the Gradio interface
+with gr.Blocks() as app:
+    gr.Markdown("### Amharic Speech-to-Text Transcription App")
+    gr.Markdown("Upload an audio file in any format, and get its transcription.")
+    
+    with gr.Row():
+        audio_input = gr.File(label="Upload Audio File", type="filepath")
+        transcription_output = gr.Textbox(label="Transcription")
+    
+    transcribe_button = gr.Button("Transcribe")
+    transcribe_button.click(transcribe_audio, inputs=audio_input, outputs=transcription_output)
+
+# Launch the app
+if __name__ == "__main__":
+    app.launch()

requirements.txt

@@ -0,0 +1,4 @@
+transformers
+speechbrain
+gradio
+pydub