From 6af3f71510045d3e0017d90e1b33b431b3d215be Mon Sep 17 00:00:00 2001 From: talksik Date: Sat, 26 Jul 2025 17:15:14 -0700 Subject: [PATCH] add brainstorm thoughts --- ai-conversation/README.md | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/ai-conversation/README.md b/ai-conversation/README.md index 29daba2..a3d484a 100644 --- a/ai-conversation/README.md +++ b/ai-conversation/README.md @@ -10,3 +10,17 @@ This proof of concept is for implementing conversational loop with AI at a low l ## Requirements Platform: the project is meant for Linux, specifically Ubuntu desktop. + +## Implementation + +The loop should be as follows: +- create the thread that manages the deepgram websocket client +- grab audio and send over to other thread to process & send +- stream pcm packets over the websocket connection +- print inbound messages from websocket connection to stdout +- signal in stdout when there is a "final message", which signifies "turn detection" + +## Open questions +1. Can deepgram websocket stream take any format? raw pcm? +[Answer here](https://developers.deepgram.com/docs/encoding) +2. Consider using Golang to interop with c, whereby the golang uses deepgram sdk, and the c library for alsa.