Newshot

Google Rolls Out 66 Regional Voices for Gemini App on Mobile, Covering 7 Language Categories

AI summary

Google has officially launched 66 regional voices for the Gemini app on mobile platforms, covering seven language categories so users can pick voices that better match local everyday speech habits for more natural conversations with Gemini Live. The update covers American English, British English, Indian English, Hindi, Brazilian Portuguese, Indonesian, and Korean, with UK users able to choose from more than 10 regional accents including Geordie, Scouse, Brummie, Scottish, Welsh, Northern Irish, and Multicultural London English. Gemini Live product management director Neel Joshi said Google worked with native speakers to capture local speaking pace, rhythm, and clarity while avoiding exaggerated regional accents, and the voices are available for Gemini Live and text-to-speech playback, selectable in the app's voice settings or via the three-dot menu during a Gemini Live conversation — though Google does not label specific city names in the menu, and Gemini will not tell users which accent is in use even if asked.

Translation(model translation, not the source's own text)

Can reproduce local everyday speaking habits: Google launches 66 regional voices for the mobile Gemini app, covering 7 language categories

IT之家 reported on November 17 that Google announced the official launch of 66 regional voices for the mobile Gemini app, covering 7 language categories, allowing users to choose voices closer to local everyday speaking habits for natural conversations with Gemini Live.

This update covers American English, British English, Indian English, Hindi, Brazilian Portuguese, Indonesian, and Korean. Among them, users in the UK can choose more than 10 different regional accents, including the Newcastle accent (Geordie), the Liverpool accent (Scouse), the Birmingham accent (Brummie), Scottish, Welsh, and Northern Irish accents, as well as Multicultural London English.

Neel Joshi, Director of Product Management for Gemini Live, said that Google worked with native speakers to reproduce the speed, rhythm, and clarity of local everyday communication while avoiding deliberately exaggerating regional accents. In other words, Gemini will not force in a large amount of slang to imitate local speaking styles; instead, it uses more natural wording and speech rhythms that match local habits to make conversations sound more friendly. The relevant voices are also available for Gemini Live and text-to-speech playback.

Users can switch accents in the Gemini app's voice settings, or adjust them during a Gemini Live conversation via the three-dot menu in the upper right corner. However, Google does not label specific city names in the voice selection menu, and even if users ask directly, Gemini will not proactively say which accent is currently being used. Google suggests that users experience the actual voice characteristics of different dialects by asking about topics such as local food, sports events, or community locations.

Related articles

  • Google's seventh-generation self-developed Ironwood TPU delivers: an 8,192-chip cluster meets the target for DeepSeek-V3 pretraining in about 102 seconds
  • Google Pics upgrades the Nano Banana 2.1 model: more refined image generation, more accurate image editing
  • The best AI assistant for App side hustles: Google launches a series of tools targeting the most troublesome monetization step
  • Weekly AI large-model rankings: Google Gemini 4 Carbon debuts strongly at the top; domestic models hold steady in the top five of the video ranking
  • Tencent WorkBuddy international version launches the Gemini 4 Carbon large model: supports a one-million-token context, currently limited to Pro users
  • Tackling AI vulnerability discovery challenges: Google Project Zero plans to gradually shorten the vulnerability public disclosure period to 30 days

The translation is machine-generated and individual terms may be off. Nothing is rewritten or cut; the original is authoritative.

Published by IT之家
This site only summarises and aggregates; copyright stays with the original author. Headlines and summaries follow the source — where they differ, the original wins.
Read the original ↗