Anthropic Expands Claude Voice Mode With Opus, Sonnet AI Models

Claude has updated its Voice Mode to support the Opus and Sonnet AI models, expanding the feature beyond its earlier reliance on Haiku. The update also enables Voice Mode to work with connected services, allowing users to complete tasks across supported apps while speaking with the chatbot. According to Anthropic's updated support page, the feature is available in beta on the web, desktop and mobile apps for all subscription plans, although some capabilities depend on the user's plan.

Claude Voice Mode Gets Opus, Sonnet Support, Gmail and Calendar Integration

The updated support page states that Voice Mode automatically starts with the latest generation of the Claude model users most recently selected in text chat. Users can switch between Opus, Sonnet and Haiku during a conversation, depending on their subscription plan. Claude Fable is currently unavailable in Voice Mode.

Previously, Voice Mode relied exclusively on the Haiku model, which prioritised quicker responses. The addition of Opus and Sonnet broadens its capabilities for longer discussions and more demanding tasks, such as brainstorming ideas, reviewing presentations and providing communication feedback.

With the update, voice conversations can access Gmail, Google Calendar, Google Docs, Slack and web search, allowing users to retrieve information or complete tasks without switching back to text. Users can check emails, review schedules, retrieve documents and perform web searches without leaving a voice conversation. They can also switch between voice and text while retaining the same chat context. Free users can connect one service, whereas paid plans support more. Anthropic notes that some tool results may not appear on screen, and using multiple connected services together could briefly increase response times.

The rollout comes weeks after OpenAI updated ChatGPT's Voice Mode with its latest conversational AI models, as competition in AI-powered voice assistants continues to grow.

Voice Mode first launched last year with support limited to the Haiku model. Earlier this year, Anthropic expanded language support beyond English by introducing beta support for French, German, Hindi, Indonesian, Italian, Japanese, Korean, Portuguese and Spanish. Users must manually select their preferred voice language from the settings menu, as changing the application's display language does not automatically update Voice Mode.

The feature is currently available in beta on Claude Mobile for Android and iOS, Claude Desktop and the web. Anthropic recommends using it on smartphones, and voice conversations count towards the usage limits of Free, Pro, Max, Team and Enterprise plans.

Voice Mode offers both hands-free and push-to-talk options. While hands-free mode responds after natural pauses in speech, push-to-talk gives users more control over when Claude listens, making it better suited to noisy surroundings. Users can also choose from several built-in voices across the desktop and mobile apps.

Anthropic has also introduced safeguards to reduce misuse. Instead of allowing custom voice generation, Voice Mode is limited to a selection of preset voices to help prevent impersonation. The company says responses are generated by Claude rather than imitating a person's speech, and the feature follows the same safety standards and usage policies as text conversations.

Voice conversations are automatically saved as text transcripts in chat history. Unlike conventional speech-to-text tools, Voice Mode is intended for back-and-forth spoken interactions instead of serving only as a transcription feature. The feature is currently unavailable in Claude Cowork and Claude Code.