KEY POINTS
- Google has added 66 regional dialects to the Gemini apps on Android and iOS.
- UK options include Geordie, Scouse, Brummie, Scots, Welsh, Northern Irish and Multicultural London English.
- Google says it worked with native speakers so the voices sound authentic rather than stereotyped.
Google has started rolling out 66 regional dialects for Gemini, letting the assistant speak to users in accents tied to specific places. Engadget reported that the feature is available from November 16, 2026, in the Gemini apps for Android and iOS.
The company first teased the capability at Google I/O earlier this year. The aim, according to the report, is to help the assistant relate better to the people using it, rather than answering in a generic, neutral voice.
In the UK, users can set Gemini to speak in Geordie, Scouse or Brummie accents, which correspond to Tyneside, Merseyside and Birmingham. The list also covers Scots, Welsh, Northern Irish and Multicultural London English, among others. The selection is not limited to Britain, as tailored dialects are also offered for India, Brazil, Indonesia and Korea.
Google says it worked with native speakers to make the voices sound authentic and clear instead of exaggerated or stereotypical. The company also said Gemini will not start filling its answers with street slang. The goal is for the assistant to sound more like a local.
Engadget noted that conversational AI can have a flattening effect on distinctive dialects and accents, since assistants often default to a uniform, neutral way of speaking. It added that, because many people take pride in their accent as a marker of heritage, it will be interesting to see how the new voices are received by the communities they are meant to represent.
To turn the voices on, users need to open the voice settings in the Gemini app or tap the three-dot menu. Google suggests one way to judge how authentic a voice sounds is to ask Gemini for a local food recommendation, which should bring out what the company describes as the warmth and rhythm of each voice.