Radio station ID and DJ voice generator
An internet radio station needs a voice between the songs: the station name, the show, the website, the time and the weather. Hiring a voice artist for every new liner is slow, and inserts that change every hour cannot be recorded in advance at all.
SuperClonica gives your station a voice for short lines, free to test online, and a free engine that voices fresh inserts on your own computer. This guide shows which voices fit radio, how to automate hourly inserts and what to keep in mind about rights.
What a station voice is for
Short lines that repeat often are where an AI voice works best.
- Station IDs and sweepers between songs.
- Liners: the show name, the website, the request line.
- Time, weather and news inserts written fresh every hour.
- Promos for your own shows and events.
Voice a station ID free
On the page of any voice, type your station ID, 30 to 100 characters, choose the language and press Voice it. In a few seconds you can listen and download it, with no account.
Voices that fit radio
Most catalog voices were recorded in Russian and keep a slight accent in other languages; for other languages, listen first or use the ten stock voices of the model.
- Energetic Narrator: a mid-low male voice with clear emphasis, for promos and liners.
- Reporter: a mid-low male voice with clear emphasis, for news inserts.
- Pleasant Announcer: low and restrained, for calm IDs and information.
- Energetic Katya: the voice of a pleasant, energetic young woman.
Fresh inserts every hour
Install the free engine: python -m pip install superclonica-cpu. A small script writes the text of the insert, with times and temperatures in words, and voices it: superclonica --text "It is three o’clock. Sunny and twenty degrees in the city." --voice M2 --lang en --normalize --output insert.mp3. The --normalize option brings the voice to −18 LUFS, so it sits evenly next to the music, and a short insert is ready in seconds on an ordinary processor. Import the files into your automation software as liners or jingles, and add music beds and effects there.
Rights and honesty
For a commercial station, use a voice you have the rights to: the ten stock voices of the model, M1 to M5 and F1 to F5, or a voice trained in the SuperClonica studio from a presenter who agreed. Character and parody voices from the catalog are for fun and fan projects. Do not imitate real presenters, and do not present a synthetic voice as a live host where listeners would care.