File size: 1,471 Bytes
1e44bd2 e395b3d d33dce2 73e0bb9 7843a47 d33dce2 73e0bb9 d33dce2 73e0bb9 d33dce2 73e0bb9 28251cb 73e0bb9 3e4448a 73e0bb9 3e4448a 73e0bb9 3e4448a 73e0bb9 3e4448a 73e0bb9 3e4448a 73e0bb9 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 |
import gradio as gr
from transformers import pipeline
# IMPORTANT: Replace this with the exact ID of your uploaded model
MODEL_ID = "Akashpb13/xlsr_kurmanji_kurdish" # Assuming your model ID uses your Space's username
# Load the ASR model pipeline
# The pipeline handles downloading the weights and configuration.
try:
transcriber = pipeline(
"automatic-speech-recognition",
model=MODEL_ID,
# device=0 # Uncomment this if you upgrade your Space to a GPU
)
except Exception as e:
# Fallback for error handling if the model fails to load
gr.Warning(f"Failed to load model: {e}")
transcriber = None
# Define the prediction function
def transcribe_audio(audio_file_path):
if audio_file_path is None:
return "Please provide an audio input."
if transcriber is None:
return "Error: Model failed to initialize."
# Pass the local file path provided by Gradio to the pipeline
result = transcriber(audio_file_path)
return result["text"]
# Create the Gradio interface
demo = gr.Interface(
fn=transcribe_audio,
inputs=gr.Audio(
sources=["microphone", "upload"],
type="filepath",
label="Kurmanji Audio Input"
),
outputs=gr.Textbox(label="Kurmanji Transcription Result"),
title="Kurmanji ASR Demo",
description="Automatic Speech Recognition for Kurmanji using a fine-tuned Hugging Face Transformer model."
)
# Launch the application
demo.launch() |