---
format: "aidr-story-markdown/v1"
id: "b787af981845bf203f2868a67d480444778c6eb533f385e00c17e409ee214460"
canonical_url: "https://aidr.today/b787af98?lang=en"
title: "Gradium Open-Sources Phonon, a 31M-Parameter On-Device TTS Model Under Apache 2.0"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-22T16:56:12.000Z"
category: "Open Source"
topics: ["open-source"]
source_urls: ["https://www.marktechpost.com/2026/10/22/gradium-phonon-31m-on-device-tts/"]
summary: "Gradium has open-sourced Phonon, a family of on-device text-to-speech models, including an English version with 31 million parameters that runs under Apache 2.0, fits in 64.5 MB, and begins speaking in about 7 ms on a laptop CPU."
---

# Gradium Open\-Sources Phonon, a 31M\-Parameter On\-Device TTS Model Under Apache 2\.0

> [Open the canonical story](<https://aidr.today/b787af98?lang=en>)

**Published:** 2026-10-22T16:56:12.000Z
**Category:** Open Source
**Topics:** open\-source

## Summary

Gradium has open\-sourced Phonon, a family of on\-device text\-to\-speech models, including an English version with 31 million parameters that runs under Apache 2\.0, fits in 64\.5 MB, and begins speaking in about 7 ms on a laptop CPU\.

## Sources

- [Story source](<https://www.marktechpost.com/2026/10/22/gradium-phonon-31m-on-device-tts/>)