Piperless

Audio transcripts for WordPress, running on your own server

Every WordPress post can be a listenable audio transcript. Piperless generates them automatically using local neural text-to-speech, no API keys, no subscriptions, no data leaving your server.

It runs on the Piper TTS engine. When you publish a post, Piperless reads the content, converts it to natural speech, and embeds a player. The text stays on your machine. The audio stays on your machine. No third-party services involved.

Why it matters

• Visually impaired readers can listen instead of read.
• Dyslexic readers get an alternative format.
• Commuting readers can consume posts as audio.
• Non-native speakers hear correct pronunciation.

What it does

• Automatic generation on publish. No per-post action required.
• 6 player themes so the audio player matches your site design.
• Per-post overrides to customize voice, language, quality, and placement.
• Multi-language support with 20+ languages via Piper voice models.
• MP3 and Opus encoding with automatic conversion from Piper's WAV output.
• Content-addressed caching so identical text never regenerates.

Why it exists

Accessibility features on the web are often bolted on or routed through third-party services that cost money and compromise privacy. I wanted a plugin that added audio transcripts to WordPress without sending content to a cloud API, without monthly fees, without privacy trade-offs.

Piper runs the TTS locally. The plugin wraps it into a polished WordPress experience with a Gutenberg sidebar, admin panel, and frontend player that doesn't look like an afterthought.

MIT License · GitHub · Releases