VoxLocal: privacy-first audio editions for indie writers
Turn your newsletter, blog, or book draft into a broadcast-quality audio edition on your own laptop — no cloud subscription, no character fees, no audio engineer.
Independent writers, Substack/Medium publishers, and educators who want to offer audio versions of their writing
- One-click import from Substack, Medium, Ghost, Notion, or pasted Markdown
- Pick a consistent narrator across issues from 50+ voices (Kokoro), with speed and emphasis controls
- One-click export to a hosted RSS feed or podcast platforms (Spotify, Apple, Pocket Casts) with proper enclosures
- Background rendering queue so writing the next post doesn't pause the last episode's render
Local CPU TTS is finally good enough that a creator can produce a credible audio edition without paying ElevenLabs per character or uploading drafts to a cloud API — a 2026 inflection point
Hacker News Kokoro post hit 241 points and Substack ships built-in free TTS read-aloud for posts, showing real indie-writer appetite for audio editions — but the built-in feature already absorbs the most casual demand.New read-aloud voices, quote restacks, and Note stats ↗100% Local, CPU-Friendly, High-Quality TTS with Kokoro ↗
Free/open tools (LM Studio, Kokoro CLI, LocalAI Pocket TTS, ElevenLabs free tier) cover the raw TTS need, but none package a polished, writer-focused audio-edition workflow (newsletter/blog/book ingest → broadcast output); positioning is possible but middle of a crowded comparison list.Local TTS Guide 2026 — Best Open-Source Text-to-Speech Models ↗12 Best ElevenLabs Alternatives in 2026 (Free & Local) ↗
Indie writers and newsletter operators are notoriously price-sensitive, Substack offers free built-in TTS, and most competing tools are free or freemium; a one-time paid desktop app faces heavy downward pricing pressure.9 Free ElevenLabs Alternatives That Actually Work in 2026 ↗Speaking up: Automatic Text to Speech voiceovers for Substack posts ↗
Local CPU TTS is on a clear quality curve (Orpheus, Kokoro, Piper) and privacy/no-cloud is a durable concern for writers with unpublished work, keeping the thesis strong for 3-5+ years.Local TTS Guide 2026 — Best Open-Source Text-to-Speech Models ↗
Mature open-source TTS models (Kokoro, Piper, XTTS, Orpheus) plus LM Studio's local API mean a writer-facing wrapper around model download → ingest text → export WAV/MP3 is well within scope for a small team.Local TTS Guide 2026 — Best Open-Source Text-to-Speech Models ↗GitHub - nazdridoy/kokoro-tts ↗