Anthropic is broadening what Claude can do by voice. Voice mode, which had been limited to Haiku, is now available for the company’s Opus and Sonnet models, giving users access to deeper analysis and more complex responses in spoken conversations.
The change also pushes voice mode beyond the chat window. Anthropic is extending access into apps like Gmail, Slack, and Canva, making voice a more practical way to move from discussion to action.
Why Opus and Sonnet matter for voice mode
When Anthropic launched voice mode in 2025, the emphasis was speed. Haiku, the model available in voice mode until now, was suited to quick questions and fast responses with minimal delay.
That design made sense for casual use. But according to Anthropic, people quickly started using voice mode for heavier tasks than simple back-and-forth questions. They were working "through real business problems," a use case that demanded more depth than Haiku was meant to provide.
Anthropic described Haiku as a model that "kept conversations quick, but not always deep." That distinction is the center of the update. Voice is no longer just a faster way to ask Claude something simple. With Opus and Sonnet added, it becomes a way to tackle work that needs more reasoning, synthesis, and follow-through.
From quick answers to harder problem-solving
Sonnet and Opus are presented by Anthropic as deeper models, "designed for hard problem-solving." In practical terms, that means Claude voice mode can now support more complex responses and analysis when a conversation requires more than a short answer.
The source examples show the direction Anthropic is taking. A user could have a conversation with Claude and then ask it to turn that discussion into a one-page pitch. Claude could also take action on behalf of the user, such as shifting calendar appointments if a train is running late.
Those examples matter because they show voice mode moving into workflow territory. A spoken exchange can become a draft, a plan, or an action across connected apps. The value is not only that the user can talk instead of type. It is that the conversation can keep going until it produces a useful output.
Switching between models and modes
Anthropic is also making the experience more flexible. Users can move between text and voice mode in the middle of a conversation. They can also change models while the conversation is already underway.
That model switching could make Claude voice mode more adaptable to different levels of work. A quick exchange can begin with Haiku, then move to Opus if the user wants to develop the idea further. The same conversation can shift from fast exploration to deeper analysis without starting over.
This is a notable change in how voice AI can fit into daily use. Voice does not have to be a separate mode for lightweight questions. It can become one way to move through a larger task, with text available when it is more convenient and a stronger model available when the work becomes more demanding.
More languages, more app reach
Anthropic is also expanding Claude voice mode to several new languages. Until now, languages other than English were only available in beta. Voice mode is now available in French, German, Spanish, Hindi, Indonesian, Italian, Japanese, Korean, and Portuguese.
That language expansion widens the potential audience for Claude voice mode beyond English users. It also makes the app integrations more meaningful, because voice access across Gmail, Slack, and Canva becomes more useful when people can speak in more languages.
The update brings together three separate shifts:
- More capable models: Opus and Sonnet join Haiku in voice mode.
- More places to act: Voice mode reaches apps like Gmail, Slack, and Canva.
- More languages: French, German, Spanish, Hindi, Indonesian, Italian, Japanese, Korean, and Portuguese are now supported.
Anthropic also issued a correction tied to the timeline of the feature. The story originally stated that voice mode launched in 2026, but it launched in 2025. That correction is important because it frames this update as an expansion of an existing feature, not the debut of voice mode itself.
The broader message is clear: Claude voice mode is becoming less of a quick-answer feature and more of a working interface for AI. By adding Opus and Sonnet, Anthropic is giving voice conversations access to models built for deeper tasks, while the app and language expansions make those conversations more useful in everyday work.