talk-llama : add an up-to-date demo video

This commit is contained in:
Georgi Gerganov 2023-11-02 15:28:48 +02:00 committed by GitHub
parent bce49a260e
commit 74de25158e
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

View File

@ -4,6 +4,10 @@ Talk with an LLaMA AI in your terminal
[Demo Talk](https://user-images.githubusercontent.com/1991296/228024237-848f998c-c334-46a6-bef8-3271590da83b.mp4) [Demo Talk](https://user-images.githubusercontent.com/1991296/228024237-848f998c-c334-46a6-bef8-3271590da83b.mp4)
*Latest perf as of 2 Nov 2023 using Whisper Medium + LLaMA v2 13B Q8_0 on M2 Ultra:*
https://github.com/ggerganov/whisper.cpp/assets/1991296/d97a3788-bf2a-4756-9a43-60c6b391649e
## Building ## Building
The `talk-llama` tool depends on SDL2 library to capture audio from the microphone. You can build it like this: The `talk-llama` tool depends on SDL2 library to capture audio from the microphone. You can build it like this: