• Sesame Launches Conversational Voice AI iOS App — The Oculus founders' startup released its public iOS app featuring four distinct personality-driven voice agents with memory, web search, and reminders across 39 countries, following over a million users in its research preview. (techcrunch.com, techcrunch.com)

  • Arc Raises $10.76M Seed for Drive-Through Voice AI — The voice AI startup launched its drive-through platform and secured a seed round led by Andreessen Horowitz, signaling early institutional conviction in purpose-built restaurant voice AI. (wsgr.com)

  • ElevenLabs Music V2 Switches Genres Mid-Track — The updated music generation model handles vocal complexity and composition across radically different genres, arriving nearly 10 months after the company's first music model. (techcrunch.com)

  • Microsoft Preparing MAI Voice 2 And Transcribe Models — Reports indicate Microsoft is readying a multilingual emotional voice model and MAI Transcribe 1.5 for a June 2 announcement. (x.com)

    BUILD 🔥: Microsoft is preparing new image and voice models for the announcement on June 2.

    > MAI Voice 2, a multilingual model supporting 15 news languages and a wider range of emotional spectrum (check voice samples in the article)

    > MAI Transcribe 1.5, a new model for… pic.twitter.com/kfiPEP6YoZ

    — 🚨 AI News | TestingCatalog (@testingcatalog) May 30, 2026

  • Claude Voice Mode Expands To 18 New Languages — Anthropic upgraded Claude's voice mode with 18 additional languages, on-the-fly language switching, new voices, UI updates, and push-to-talk functionality. (x.com)

    ANTHROPIC 🔥: Voice mode on Claude mobile apps is about to get an upgrade with 18 new supported languages!

    > Claude will be able to change language on the fly
    > All languages have 1-2 new voices
    > Voice Mode UI will get a new look
    > A new push-to-talk functionality will be… pic.twitter.com/rUes3rXYdG

    — 🚨 AI News | TestingCatalog (@testingcatalog) May 27, 2026

  • Alibaba TTS Model Reaches Global Top Five — Alibaba's Fun-Realtime-TTS-Preview from Tongyi Lab reached an Elo score of 1190 on the Artificial Analysis Speech Arena leaderboard, ranking fifth globally and first among Chinese models. (scmp.com)

  • Rime Coda TTS Now Available On Telnyx — Rime's new flagship TTS model, the successor to Arcana, is now accessible through the Telnyx platform, expanding enterprise distribution for the voice model. (telnyx.com)

  • Apple Previews Major Siri Overhaul For iOS 27 — Bloomberg reports a revamped Siri interface and new chatbot-style app among the major changes planned for announcement at Apple's June 8 Worldwide Developers Conference. (bloomberg.com)

  • Google Gemini Spark Debuts As 24/7 Agentic Assistant — First introduced at Google I/O, the new always-on assistant is designed to handle inbox management, task automation, and digital life organization through conversational interaction. (techcrunch.com)

  • ASR Bias Research Flags Career Risk For Non-Western Names — Emerging evidence shows that AI transcription and meeting-analytics tools from platforms like Granola and Fireflies can misattribute or drop contributions from employees with non-Western names, creating measurable career disadvantages. (forbes.com)

  • Research Highlights Psychosocial Differences In Voice Vs Text AI — New findings reveal counterintuitive psychosocial outcomes when users interact with AI in voice mode compared to text, with implications for how conversational voice agents should be designed and deployed. (forbes.com)

  • Pennsylvania Seeks Injunction Against Chatbot Claiming Medical License — The state filed legal action against an AI maker whose chatbot persona claimed to be a licensed psychiatrist, raising urgent questions about persona guardrails in voice and conversational AI products. (forbes.com)