Spaces:
Running
Running
talk-llama : add an up-to-date demo video
Browse files
examples/talk-llama/README.md
CHANGED
|
@@ -4,6 +4,10 @@ Talk with an LLaMA AI in your terminal
|
|
| 4 |
|
| 5 |
[Demo Talk](https://user-images.githubusercontent.com/1991296/228024237-848f998c-c334-46a6-bef8-3271590da83b.mp4)
|
| 6 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
## Building
|
| 8 |
|
| 9 |
The `talk-llama` tool depends on SDL2 library to capture audio from the microphone. You can build it like this:
|
|
|
|
| 4 |
|
| 5 |
[Demo Talk](https://user-images.githubusercontent.com/1991296/228024237-848f998c-c334-46a6-bef8-3271590da83b.mp4)
|
| 6 |
|
| 7 |
+
*Latest perf as of 2 Nov 2023 using Whisper Medium + LLaMA v2 13B Q8_0 on M2 Ultra:*
|
| 8 |
+
|
| 9 |
+
https://github.com/ggerganov/whisper.cpp/assets/1991296/d97a3788-bf2a-4756-9a43-60c6b391649e
|
| 10 |
+
|
| 11 |
## Building
|
| 12 |
|
| 13 |
The `talk-llama` tool depends on SDL2 library to capture audio from the microphone. You can build it like this:
|