SUB/WAVE
A real internet radio station.
Is this your product? Claim it to edit the listing, post updates and answer questions. Prove it with an email address at the product's domain (the listing becomes Claimed), or put the LotsDirectory badge on your website (the listing becomes Verified, and its link to your website is followed while the badge is live).
Description
A real internet radio station. Single Icecast stream, every listener hears the same broadcast at the same time, picked and announced by an LLM-driven DJ.
About SUB/WAVE
What is SUB/WAVE?
SUB/WAVE is an open source internet radio station platform. It broadcasts a single shared Icecast audio stream where tracks and station announcements are chosen and spoken by an autonomous AI DJ.
What SUB/WAVE does
The system runs on a single machine using four processes: a Node.js controller, an LLM brain, Liquidsoap for audio mixing, and Icecast to distribute the audio stream in MP3, Opus, AAC, or FLAC formats. The AI DJ selects tracks from a local music library, announces song intros, gives time and weather checks, and handles up to twenty-four distinct persona identities with voice cloning. Listeners tune into the same live stream via web, desktop, or mobile applications, with lock screen media controls and custom visual skins. A listener request drawer accepts natural language song requests, which the LLM matches to library tracks to introduce on air. An admin console allows the operator to brush shows onto a 24x7 grid, tag library moods, and review system diagnostics.
Who SUB/WAVE is for
It is built for radio operators and listeners who want to run a personal radio station with an AI DJ and their own music library.
How SUB/WAVE works
The controller checks schedule rules, weather, and listener requests, then prompts a language model to select the next track and write spoken links. A speech engine synthesizes the speech, Liquidsoap crossfades the audio with jingles and music, and Icecast transmits the synchronized broadcast to all connected listeners.
Key features
- Single Icecast stream where every listener hears the same broadcast at the same time
- Autonomous AI DJ selecting tracks and generating spoken voice links between songs
- Persona system supporting up to twenty-four DJ identities with voice cloning and guest co-hosts
- Natural language request drawer matching listener song requests against the music library
- Local or cloud speech synthesis integration supporting Piper, Kokoro, Chatterbox, and ElevenLabs
- Four processes on one box: Node.js controller, LLM brain, Liquidsoap mixer, and Icecast streaming
- Multi-platform listening support across web PWA, native desktop, iOS, Android, and M3U playlists
- Administrative console for managing 24x7 show schedules, library mood tags, and station health
Facts from getsubwave.com, checked 14 October 2026
Written from SUB/WAVE's own website (Homepage). The maker can claim this listing to edit it.
Categories
Frequently asked questions
What audio formats does SUB/WAVE support for streaming?
The platform streams through Icecast using MP3 by default, with optional Opus, AAC, and lossless FLAC mounts.
Can listeners skip songs on SUB/WAVE?
No. The station operates as a shared broadcast where every listener hears the same stream simultaneously with no skip button.
Which speech synthesis engines can be used for the DJ voice?
Supported speech engines include local options like Piper, Kokoro, Chatterbox, and PocketTTS, as well as cloud providers like OpenAI and ElevenLabs.
How does the listener song request system operate?
Listeners type requests in natural language, and the LLM searches the library to queue matching tracks for the DJ to announce on air.