SubVox: a native GPU text-to-speech engine in .lx

SubVox is a native Windows text-to-speech engine written in .lx. A single 0.4 MB executable drives the GPU through Direct3D 11 to run an open 31M English speech model, producing 24 kHz speech and a timestamp for every word, with no Python, runtime, browser or installer. Measured on an RTX 5070 Ti: 134x faster than real time, first audio in 4.5 ms. 37 voices, a native GUI, Apache 2.0. Code and release: https://github.com/Aetherion-Substrata/SubVox KIZ - SOLO DEV - Aetherion Substrata Ltd - FIZZNETWORK

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-05
DOI
https://doi.org/10.5281/zenodo.23989994
Primary Topic
Speech Recognition and Synthesis
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

SubVox: a native GPU text-to-speech engine in .lx

KIZ
Zenodo (CERN European Organization for Nuclear Research)
Speech Recognition and Synthesis
article

SubVox: a native GPU text-to-speech engine in .lx

KIZ
article en

Abstract

SubVox is a native Windows text-to-speech engine written in .lx. A single 0.4 MB executable drives the GPU through Direct3D 11 to run an open 31M English speech model, producing 24 kHz speech and a timestamp for every word, with no Python, runtime, browser or installer. Measured on an RTX 5070 Ti: 134x faster than real time, first audio in 4.5 ms. 37 voices, a native GUI, Apache 2.0. Code and release: https://github.com/Aetherion-Substrata/SubVox KIZ - SOLO DEV - Aetherion Substrata Ltd - FIZZNETWORK

Zenodo (CERN European Organization for Nuclear Research)
Openalex Percentile: Top 16%
Speech Recognition and Synthesis
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon
SubVox: a native GPU text-to-speech engine in .lx — KIZ · Zenodo (CERN European Organization for Nuclear Research) (2026) | TGRS Research Map | TGRS