News anchor voice generator: an AI voice that reads the news
A daily news digest, a company newsletter read aloud, a report for a YouTube channel, a news insert in a game or a parody news show: all of them need a voice that is clear, even and sure of itself. A news anchor voice generator reads your text in such a voice in seconds.
SuperClonica has voices made for news and explanations, a free online test and a free engine that reads long texts on your computer. This guide shows which voices sound like a newsroom, how to write copy for the ear, how to voice a whole digest offline and how to use it honestly.
News voices in the catalog
Every voice speaks the 31 languages of Supertonic 3. Most catalog voices were recorded in Russian and keep a slight accent in other languages, so test a full news item in your language first.
- Reporter: mid-low male voice with clear emphasis, for news chronicles, event reports and news inserts.
- Young Analyst: mid-low with light emphasis, reads reviews and comparisons clearly, no theatrics.
- Pleasant Announcer: low and restrained, for explanations and voice-over.
- Eva: a neutral, restrained female voice for a calm address.
- Police Radio Girl: a radio-style voice recorded in English, for a scene or a game.
- The ten stock voices of the model, M1 to M5 and F1 to F5: free in the engine.
Read a headline free online
On the page of any voice, paste a headline or the first sentence of a story, 30 to 100 characters, choose the language and press Voice it. In a few seconds you hear how your news will sound and can download the recording. No account is needed.
Write copy for the ear
Read a test item at the default speed first; if it sounds rushed, add full stops rather than slowing the whole text.
- Lead with the main fact: who, what, when, in the first sentence.
- One fact per sentence; long sentences lose the listener.
- Figures can stay as digits: the engine reads them as words in 27 languages. Check sentences with numbers in Hindi, Greek, Estonian and Croatian, where it does not.
- Write names and abbreviations the way they should be read the first time they appear.
A whole digest offline
Install the free SuperClonica engine: python -m pip install superclonica-cpu. Save each story as a text file and voice it with one command: superclonica --text-file story-01.txt --voice M2 --lang en --normalize --output story-01.wav. The engine splits long text into pieces, --normalize brings every story to the same loudness, −18 LUFS, and the result is a 16-bit WAV file for any editor. It runs offline on an ordinary processor, with no character limit, so a daily digest costs nothing extra.
Honest news
Say that the voice is generated, and never make it sound like a real broadcaster, a real channel or a real person. Do not use a news voice to pass off invented events as real reports. A parody news show is fine when it is clearly marked as parody. For your own channel’s anchor, record a clean phrase of 5 to 10 seconds of a presenter who agreed and train the voice in the studio, or order it on the exchange.