diff options
| author | Pinapelz <yukais@pinapelz.com> | 2026-04-19 19:44:14 -0700 |
|---|---|---|
| committer | Pinapelz <yukais@pinapelz.com> | 2026-04-19 19:44:14 -0700 |
| commit | 054b742d3f2a23a09209f392380f6af0f3e31dce (patch) | |
| tree | 62fee8c23326907ee6fa59ed1c50bc7c7b66d0cd | |
| parent | 8d75470ecdb4cb70a6c114adee7e931a5ea8eae5 (diff) | |
add CUDA install to README
| -rw-r--r-- | README.md | 4 |
1 files changed, 4 insertions, 0 deletions
@@ -3,6 +3,10 @@ A basic LOCAL translation backend that listens to an audio sink via PCM and runs Translations and trascriptions are transformers based, inaccuracies and hallucinations will occur. +> It's highly recommended that you run this with a GPU, running with CPU is possible but inference will be very slow outside of using tiny models (which compromise accuracy) +> +> For this, you will need to install a Nvidia CUDA 12 toolkit. I am running with [CUDA Toolkit 12.9](https://developer.nvidia.com/cuda-12-9-0-download-archive) + `server.py` serves a backend for translating incoming audio data. It expects some other client to hit the `/events` endpoint to fetch the translated data. `youtube-subtitle.user.js` is one such example client that can fetch data from this endpoint and render it beneath a YouTube video. You can install it as a userscript. |