Create app.py
Browse files
app.py
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import gradio as gr
|
| 2 |
+
from transformers import pipeline
|
| 3 |
+
|
| 4 |
+
# Load the ASR pipeline
|
| 5 |
+
pipe = pipeline(
|
| 6 |
+
"automatic-speech-recognition",
|
| 7 |
+
model="lyimo/whisper-small-sw-badili-v4"
|
| 8 |
+
)
|
| 9 |
+
|
| 10 |
+
def transcribe(audio):
|
| 11 |
+
if audio is None:
|
| 12 |
+
return ""
|
| 13 |
+
|
| 14 |
+
# Process audio file path with pipeline
|
| 15 |
+
result = pipe(
|
| 16 |
+
audio,
|
| 17 |
+
generate_kwargs={"language": "swahili"}
|
| 18 |
+
)
|
| 19 |
+
return result["text"]
|
| 20 |
+
|
| 21 |
+
# Create Gradio interface
|
| 22 |
+
interface = gr.Interface(
|
| 23 |
+
fn=transcribe,
|
| 24 |
+
inputs=gr.Audio(sources=["microphone", "upload"], type="filepath"),
|
| 25 |
+
outputs=gr.Textbox(label="Transcription"),
|
| 26 |
+
title="Swahili Speech Recognition",
|
| 27 |
+
description="Record or upload Swahili audio to see the Whisper transcription",
|
| 28 |
+
allow_flagging="never"
|
| 29 |
+
)
|
| 30 |
+
|
| 31 |
+
# Launch the app
|
| 32 |
+
interface.launch()
|