The best AI note-taking tools 👀, Updates from Deepgram, RingCentral, Zoho and much more 🔥

The best AI note-taking tools for meetings from TIME

• Circleback - top pick for an in-meeting note-taking bot

• Krisp - top pick for note-taking apps that do not use an in-meeting bot

• Granola - top pick for users preferring to manually take meeting notes

Top Updates 💪

• Deepgram acquires Poised, elevating real-time voice AI communication

• RingCentral pushes AI deeper into sales tech, integrates Teams

• OpenAI seems defensive about its AI voice engine

• Stratodesk partners with Krisp for enhanced audio clarity in secure workspaces

• Cognigy lands cash to grow its contact center automation business

• Zoho announces strategic updates across collaboration solutions

  

• Meaning announces partnership with Five9, to improve customer experiences

• Innovaccer launches AI-powered healthcare contact center

• Phonely, AI-driven phone answering service startup, raises $500K in funding

• Apple offers developers a new translation API

  

• Canary Speech, voice biomarker health tech company raises $13M in funding

• AI-powered voice clarity startup Tomato.ai launches with $2.1M in seed funding

• Record, summarize, and transcribe audio with Apple Notes on iOS 18

• Everyone wants a piece of the Contact Center QA market, and that’s not good

• AssemblyAI introduces German STT and enhances PII detection

Noteworthy 💪

• Trust AI to speak to your customers? 5 ways AI could hurt your brand voice

• The AI revolution is coming for your headphones. Are you ready?

• Whisper found to generate harmful, violent text due to hallucinations

• Apple intelligence GenAI, ChatGPT to boost Siri on iPhone, Mac

• SoftBank's new AI makes angry customers sound calm on phone

• Essential components for effective conversational AI

• What are AI agents? Artificial Intelligence agents capabilities

• AI speech-to-text can hallucinate violent language

---

Science and Demo Corner 😎

• Microsoft’s VALL-E 2: Human parity in zero-shot text-to-speech achieved

• MARS5, open-source, insanely prosodic text-to-speech model

• Coqui-ai/TTS: A deep learning toolkit for TTS

• Audio Mamba: Transformer-level performance without self-attention

• Researchers propose first ‘Transcription-free’ automatic subtitling model

• Whisper WebGPU: Real-time in-browser ASR with OpenAI Whisper

• StreamSpeech: ‘All in one’ model for STT, TTS and Speech Translation

• AudioMarkBench: Benchmarking robustness of audio watermarking

• Minimal knowledge PT-AE attacks on black-box Speaker Recognition models

• The spiking neural network based on fMRI for speech recognition

• Nonlinear regularization decoding method for speech recognition

• Here’s a new dataset for emotion-aware speech translation