The short version
A C++ audio-model inference engine recommended as an easy way to try multiple models.
From the HackerLinks archive
audio.cpp
A C++ audio-model inference engine recommended as an easy way to try multiple models.
At a glance:
First seen:2026-09-24
Last seen:2026-09-24
Times seen:1
Website:github.com
Why it caught our attention
A commenter directly endorsed its utility while comparing text-to-speech options.
Where it surfaced on Hacker News
2026-09-24
“is an easy way to run a lot of different models.”
VariousPrograms · recommendation · first hand use
Direct HN comment
Gemini 3.8 text-to-speech
Also surfaced in this discussion
pdf-narrator
Gemini 3.8 text-to-speech
2026-09-24
Pocket TTS
Gemini 3.8 text-to-speech
2026-09-24