Multilingual voice assistants are entering a new phase
A multilingual voice assistant is an AI-powered helper that can listen, understand, and respond to spoken commands or conversations in multiple languages without needing manual language changes. As Google and Anthropic roll out major upgrades, this definition is becoming more practical than theoretical. Anthropic’s Claude Voice Mode now adds multilingual controls and a push-to-talk option, while Google’s Gemini app expands voice input to more than 70 languages with flexible language mixing. At the same time, an upgraded Google Voice Search experience brings live chat, translation, and song recognition into the same interface. Together, these products show how AI language support is shifting from English-first to language-inclusive design. For millions of users, especially those who think and speak in more than one language, these updates shrink the gap between how they talk in daily life and how they interact with AI.
Claude Voice Mode adds language choice and push-to-talk control
Anthropic is rolling out its biggest upgrade yet for Claude Voice Mode, with new multilingual options and a more flexible control scheme. Users can now pick from languages such as Spanish (Latin America), German, Portuguese, Chinese, Japanese, Russian, and Ukrainian inside the app’s Voice settings. Once set, Claude can handle spoken conversations in these languages, making it a stronger multilingual voice assistant for non-English speakers. A key usability change is the new push-to-talk mode: instead of relying only on hands-free conversation, users can press and hold a button while speaking, then release it to send their message. This reduces unintended activations and gives people more precise control over when Claude listens. According to Android Authority, some testers have also spotted a phone-call-style icon in the iOS build, hinting that Anthropic may be exploring more phone-like, continuous audio sessions in future updates.
Gemini voice commands can now mix more than 70 languages
Google’s Gemini app is turning into one of the most flexible multilingual voice assistants available by dramatically expanding its mic input. Josh Woodward, Google’s VP for Gemini, said the app’s voice input now supports over 70 languages and works automatically without any manual language switching. Users can speak naturally, blending languages in a single sentence—such as starting in Hindi and inserting English phrases—and Gemini voice commands will still be understood. This feature is available on Android and iOS and is scheduled to reach the web shortly, making the same experience consistent across platforms. The update pairs with Google’s recent Gemini Omni model and Gemini 3.5 Live Translate, positioning the app as a central hub for spoken queries, translation, and AI assistance. For bilingual and multilingual users, the ability to mix languages mirrors real conversation, instead of forcing a strict one-language-at-a-time rule.
Google Voice Search evolves into a richer conversational tool
Alongside Gemini, Google is enhancing the broader voice experience with an updated Google Voice Search that now supports live chat, translation, and song recognition. This turns the familiar mic button into a more capable gateway to AI language support, going beyond simple keyword queries. Users can speak to start a live chat-like exchange, ask for on-the-fly translations, or hum and sing so the system can identify songs. These features mirror elements of dedicated AI chatbots and music recognition tools, but in a single voice-first interface. By folding more tasks into voice search, Google reduces the need to switch between different apps or modes when moving from a simple query to a longer conversation. The result is a more coherent voice experience, especially for people who rely on spoken interaction because of typing barriers or language preferences.
Why these upgrades matter for everyday multilingual users
Taken together, Claude Voice Mode’s multilingual settings, Gemini’s flexible voice commands, and Google Voice Search’s new conversation and recognition tools mark a clear shift in AI language support. Voice interfaces are no longer designed mainly around a single dominant language; they are being built for people who comfortably move between tongues. This lowers barriers for those who prefer speaking rather than typing, and for users who think in one language but search in another. Push-to-talk in Claude gives more control in noisy environments, while Gemini’s automatic language detection removes the friction of repeated settings changes. Song recognition and live translation in voice search round out the experience. As these features continue rolling out, multilingual users gain AI assistants that feel less like foreign tools and more like responsive, conversational partners that match how they already speak every day.
