Tech

Anthropic expands Claude voice capabilities to Opus and Sonnet models

The update allows users to perform complex business tasks via voice and moves nine non-English languages out of beta.

Author
Owen Mercer
Markets and Finance Editor
Published
Draft
Source: The Verge · original
Claude’s voice mode is now available for Opus and Sonnet
AI provider extends voice mode beyond Haiku, adding tool-use for Gmail, Slack and Canva

Anthropic has significantly widened the scope of its Claude voice mode, extending access to its more powerful Opus and Sonnet models. Previously, voice interaction was restricted to Haiku, a faster but less complex model designed for rapid responses with minimal delay. This latest update addresses a gap identified in early usage data, where users applied the feature to complex business problems rather than simple queries, tasks for which Haiku was not optimised.

The expansion allows users to engage in deeper conversations and execute actionable workflows through voice commands. Users can now access external applications including Gmail, Slack, Canva, Notion and Google Calendar. This integration enables practical outcomes such as drafting documents, managing calendars and converting voice discussions into one-page pitches, moving beyond the text-only limitations of earlier iterations.

Anthropic noted that while Haiku kept conversations quick, it often lacked the depth required for hard problem-solving. The newly integrated Opus and Sonnet models are designed to deliver more complex analysis and take action on behalf of the user. For example, the system can adjust calendar appointments in response to real-time updates or refine detailed arguments based on voice input.

Users retain the flexibility to switch between text and voice modes mid-conversation and can seamlessly transition between model families. A user might begin with Haiku for quick ideation and then switch to Opus for detailed analysis without breaking the flow of the interaction. The system is also designed to automatically default to the fastest available model version where appropriate.

In addition to model expansion, Anthropic has moved voice mode from beta to general availability in nine non-English languages. The feature now supports French, German, Spanish, Hindi, Indonesian, Italian, Japanese, Korean and Portuguese. This marks a significant shift from the previous English-only beta status for international users, broadening the accessibility of the tool for global enterprise and consumer applications.

Continue reading

More from Tech

Read next: Open-source tool claims 97 per cent token savings for AI agents
Read next: Valvoline Unveils August 2026 Promotional Offers for Service and Retail Buyers
Read next: Developer Antirez releases native MiniMax H3 inference engine for Apple Silicon