import gradio as gr | |
from transformers import pipeline | |
# Define the full path to the NeMo Canary model | |
model_name = "nvidia/canary-1b" | |
# Define the pipeline for speech recognition with the NeMo Canary model | |
pipe = pipeline( | |
task="automatic-speech-recognition", | |
model=model_name | |
) | |
# Create a Gradio interface from the pipeline | |
interface = gr.Interface.from_pipeline(pipe, inputs="microphone", outputs="text") | |
# Launch the interface | |
interface.launch() | |