Short answer: to make an announcer voice, write the line with Grok speech tags (<loud> for the big moments, <build-intensity> to ramp up, <emphasis> on key words and [pause] for the crowd) and generate it with Grok text to speech. The box above holds a boxing intro; the buttons load a sports call, a game show, a PA announcement and a movie trailer. Pick one of 28 voices and press Generate. Trying it needs no account.
How to make an announcer voice
- Pick an announcement with the buttons above the tags, or type your own. Each example is under 200 characters, so it plays without an account.
- Put in your names. The fighter, the team, the event. Keep the tags around the parts that should hit hardest; the tags are not read out.
- Pick a voice. All 28 voices accept the same tags. The clips above use Eve.
- Press Generate and listen. Press Download MP3 to save it; without an account the file ends with a short spoken credit.
A ring announcement that builds to the name:
<loud>Ladies and gentlemen!</loud> [pause] In the red corner, weighing in at two hundred and ten pounds, <build-intensity>the undefeated champion of the world!</build-intensity>
Which tags make an announcer
All of these are among the 26 speech tags in xAI's text to speech docs. Which tag suits which moment is our suggestion:
| Tag | What it does | Use it for |
|---|---|---|
<loud>…</loud> | A raised, pushed delivery | "Ladies and gentlemen!", "What a goal!" |
<build-intensity>…</build-intensity> | Rises over the wrapped phrase | The run-up to a name or a score |
<emphasis>…</emphasis> | Stresses the wrapped words | Prize amounts, team names |
<fast>…</fast> | Speeds the words up | Play-by-play action |
[pause], [long-pause] | A silence | Room for the crowd, a dramatic beat |
[inhale] | A breath in | Right before the big call |
<lower-pitch>…</lower-pitch>, <slow>…</slow> | Lower and slower | Movie trailer voice-over |
In our two pre-recorded clips of the same words in the same voice, the tagged one ran 8.7 seconds against 7.4, with a one-second pause after "Ladies and gentlemen!". Its average level was about 1 dB lower: the tags change how the line is delivered, not the volume of the file.
Ideas to try
These are suggestions, not tests we ran for each use:
- Boxing, MMA and wrestling intros: the example above, with your fighter's name inside
<build-intensity>. - Sports highlights and fantasy league videos:
<fast>for the play,<loud>for the finish. - Game shows, quizzes and classroom games:
<emphasis>on the prize,[pause]before the answer. - Events, school assemblies and PA systems: a calm read with
<emphasis>on "Attention, please". - Streams and YouTube intros: a short hype line to play when you go live.
Tips for a better announcement
- Write it the way it is shouted. Short phrases and exclamation marks give the line its rhythm.
- Build to one moment. One
<build-intensity>toward the name works better than shouting everything. - Leave room. A
[pause]after the opening call makes the next line land. - Try two or three voices. Our voice pages suggest Rex for ads, trailers and sports, Perseus for sports and action, and Helios for announcements and event intros (suggestions, not xAI descriptions); the voices page has every voice reading one sentence.
To shout a whole script, use the shouting text to speech page; for phone greetings, the voicemail greeting generator. Every tag is in the speech tags guide.
Limits, price and privacy
Signed out, you get 3 previews a day of up to 200 characters, downloadable with a short spoken credit at the end; a free account removes the credit, and a Google account raises the limit to 1,000 characters per clip. The tag characters count toward the limit. Your text goes to xAI's model through our Cloudflare function to make the audio and is not stored.
If you build your own app, the same model costs $15 per million characters through xAI's API; see the pricing calculator. There is no separate price for an announcer voice; xAI bills text to speech per character.
Last updated: