Create app.py
Browse files
app.py
ADDED
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import gradio as gr
|
2 |
+
from transformers import pipeline
|
3 |
+
from fpdf import FPDF
|
4 |
+
import librosa
|
5 |
+
|
6 |
+
def transcribe_and_generate_pdf(audio_file):
|
7 |
+
transcriber = pipeline("automatic-speech-recognition", model="openai/whisper-large")
|
8 |
+
audio, _ = librosa.load(audio_file, sr=16000)
|
9 |
+
transcription = transcriber(audio)["text"]
|
10 |
+
|
11 |
+
output_pdf = "transcription.pdf"
|
12 |
+
pdf = FPDF()
|
13 |
+
pdf.add_page()
|
14 |
+
pdf.set_font("Arial", size=12)
|
15 |
+
pdf.multi_cell(0, 10, transcription)
|
16 |
+
pdf.output(output_pdf)
|
17 |
+
|
18 |
+
return transcription, output_pdf
|
19 |
+
|
20 |
+
interface = gr.Interface(
|
21 |
+
fn=transcribe_and_generate_pdf,
|
22 |
+
inputs=gr.Audio(source="upload", type="filepath"),
|
23 |
+
outputs=[gr.Textbox(label="Transcription"), gr.File(label="Download PDF")],
|
24 |
+
title="Audio-to-Text and PDF Generator",
|
25 |
+
)
|
26 |
+
|
27 |
+
if __name__ == "__main__":
|
28 |
+
interface.launch()
|