make it possible to record audio from within the app

app.py

@@ -2,6 +2,9 @@ import gradio as gr
 from speechbrain.inference.ASR import EncoderASR
 from pydub import AudioSegment
 import os
+import numpy as np
+from scipy.io.wavfile import write
+
 
 # Load the ASR model
 asr_model = EncoderASR.from_hparams(
@@ -31,13 +34,32 @@ def transcribe_audio(audio_file):
     
     return transcription
 
+def save_audio_to_file(audio_data, file_path="temp_audio/input_audio.wav"):
+    """
+    Converts Gradio audio data (numpy array and sample rate) to a .wav file.
+    """
+    audio_array, sample_rate = audio_data
+    # Normalize and save as .wav
+    os.makedirs(os.path.dirname(file_path), exist_ok=True)
+    write(file_path, sample_rate, np.int16(audio_array * 32767))  # Convert float32 to int16
+    return file_path
+
+def process_audio(audio_data):
+    """
+    Processes recorded/uploaded audio, saves it, and sends it to the transcribe_audio function.
+    """
+    temp_audio_path = save_audio_to_file(audio_data)  
+    transcription = transcribe_audio(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.")
+    gr.Markdown("Upload or record an audio file in any format, and get its transcription.")
     
     with gr.Row():
-        audio_input = gr.File(label="Upload Audio File", type="filepath")
+        audio_input = gr.Audio(label="Upload or Record Audio", type="filepath")
         transcription_output = gr.Textbox(label="Transcription")
     
     transcribe_button = gr.Button("Transcribe")