Updates from Google, MS, Twilio, Alibaba, Genesys and more 🔥

Voice AI weekly digest

Top Updates 💪

• Google’s Gemini AI can now analyse your audio files (Business Standard)

• Microsoft’s Copilot AI TTS gets new, cleaner “scripted mode” (PCWorld)

• Twilio adds OpenAI API for seamless AI voice warm transfers (WebProNews)

• Alibaba unveils its most powerful speech model (36Kr Europe)

• SoundHound AI acquires Interactions for $60MN (CX Today)

• Genesys deepens its ServiceNow partnership, releases new agentic AI orchestration capabilities (CX Today)

• Podonos raises $2.4M in pre-seed funding (FINSMES)

• Verbit and Deepdub partner to automate multilingual dubbing (PR Newswire)

• Milagro and Revmo AI partner to transform restaurant guest engagement with conversational voice AI (PR Newswire)

• Intella raises $12.5M to offer AI STT for 25+ Arabic dialects (MENAbytes)

• TwinMind raises $5.7M to launch AI second brain for note-taking (Dataconomy)

• AiOla brings voice AI to complex industries – with Snowflake (Snowflake)

• Why big investors are all ears for voice AI startups (Crunchbase)

• Ralph Lauren begins rollout of AI conversational shopping experience (PYMNTS)

• Deepdub launches Lightning, a real-time voice model (Yahoo Finance)

• Ai phone call assistant: Merging human touch with voice AI (London Daily News)

• How AI transcription is revolutionizing business communication (TechBullion)

• Bolna AI bets on powering all voice models (Analytics India Magazine)

• RingCentral acquires CommunityWFM to expand RingCX portfolio (RingCentral)

• CallTrackingMetrics launches VoiceAI for contact centers (PR Newswire)

• Startup Hello Patient raised $22.5 million in series A (FierceHealthcare)

• AI voice agents helped improve accuracy of blood pressure measurements in older adults (News-Medical)

Voice AI Podcast 🎙️

In case you missed the latest episode of Voice AI Podcast…

---

Engineering Corner 😎

• Best TTS chrome extensions (AboutChromebooks)

• OLMoASR: a series of open automatic speech recognition models (AllenAI)

• Dual-stream former: A dual-branch transformer architecture for visual speech recognition (MDPI)

• Building a speech-enhancement and automatic speech recognition pipeline in Python using SpeechBrain (MarkTechPost)

• The future of conversational voice AI is here by Pannalabs.ai (DEV)

• Voice recognition vs speech recognition: What you need to know (ClickUp Blog)

• Stanford & UC Santa Cruz launch benchmark for audio-language models (Slator)

• News Corp Australia rolls out TTS audio innovation (Newscorp Australia)

  ---

Fullband 2025 registration is open🔔

Join us for Fullband 2025, Krisp’s annual voice innovation conference, where CX leaders, analysts, and innovators explore the role of voice in customer contact and the technologies shaping its future.