nvidia/canary-qwen-2.5b

Update app.py

Update app.py

Update app.py

Update requirements.txt

Update requirements.txt

Update app.py

reorder

fixes

Add LLM capabilities in the demo

Reduce peak GPU memory use during model init

Fix name

Show segment timestamps for long audio

bump gradio ver

Works

Adapt to Canary-Qwen-2.5B

Update gradio sdk_version to 5.14.0 (#4)

update requirements.txt with commit from r2.1.1 and use new syntax

Update README.md (#3)

update path to chunked infer script in main branch

add text about audio duration in step 1

use main branch commit from after buffered inference merge

librosa load mono = True

allow buffered inference up to 10 mins

Update app.py

Create requirements.txt

Create pre-requirements.txt

Create packages.txt

Update README.md

Create app.py

initial commit