Create app.py
Browse files
app.py
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import gradio as gr
|
| 2 |
+
import speech_recognition as sr
|
| 3 |
+
from pydub import AudioSegment
|
| 4 |
+
|
| 5 |
+
# Initialize the recognizer
|
| 6 |
+
recognizer = sr.Recognizer()
|
| 7 |
+
|
| 8 |
+
def transcribe_audio(audio):
|
| 9 |
+
try:
|
| 10 |
+
# Convert audio input to .wav if needed
|
| 11 |
+
audio_data = AudioSegment.from_file(audio.name)
|
| 12 |
+
audio_data.export("audio.wav", format="wav")
|
| 13 |
+
|
| 14 |
+
# Use recognizer to transcribe the audio
|
| 15 |
+
with sr.AudioFile("audio.wav") as source:
|
| 16 |
+
audio_content = recognizer.record(source)
|
| 17 |
+
transcription = recognizer.recognize_google(audio_content)
|
| 18 |
+
|
| 19 |
+
return transcription
|
| 20 |
+
except Exception as e:
|
| 21 |
+
return f"Error: {str(e)}"
|
| 22 |
+
|
| 23 |
+
# Create the Gradio interface
|
| 24 |
+
with gr.Blocks() as demo:
|
| 25 |
+
gr.Markdown("# Lecture Note Transcription App")
|
| 26 |
+
|
| 27 |
+
with gr.Row():
|
| 28 |
+
audio_input = gr.Audio(source="microphone", type="filepath", label="Record Lecture Audio")
|
| 29 |
+
output_text = gr.Textbox(label="Transcription", lines=10)
|
| 30 |
+
|
| 31 |
+
submit_button = gr.Button("Transcribe")
|
| 32 |
+
|
| 33 |
+
# Link the transcribe function to Gradio components
|
| 34 |
+
submit_button.click(fn=transcribe_audio, inputs=[audio_input], outputs=[output_text])
|
| 35 |
+
|
| 36 |
+
# Launch the Gradio app
|
| 37 |
+
demo.launch()
|