---
format: "aidr-story-markdown/v1"
id: "c7641f54401d6258713d349166cc51567373b50a854ee481aba2eb19d1a51d13"
canonical_url: "https://aidr.today/c7641f54?lang=en"
title: "A game scripting language that runs speech-to-text 5x faster than llama.cpp on CPU"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-31T19:52:22.000Z"
category: "Frameworks"
topics: ["inference","framework"]
source_urls: ["https://dasllama.io/articles/2026-10-28-the-need-for-speed.html","https://lobste.rs/s/zhyenw/game_scripting_language_runs_speech_text"]
summary: "A game scripting language has been developed that runs speech-to-text 5x faster than llama.cpp on CPU. The implementation involved writing an inference engine and associated macros."
---

# A game scripting language that runs speech\-to\-text 5x faster than llama\.cpp on CPU

> [Open the canonical story](<https://aidr.today/c7641f54?lang=en>)

**Published:** 2026-10-31T19:52:22.000Z
**Category:** Frameworks
**Topics:** inference, framework

## Summary

A game scripting language has been developed that runs speech\-to\-text 5x faster than llama\.cpp on CPU\. The implementation involved writing an inference engine and associated macros\.

## Sources

- [Story source](<https://dasllama.io/articles/2026-10-28-the-need-for-speed.html>)
- [Discussion](<https://lobste.rs/s/zhyenw/game_scripting_language_runs_speech_text>)