Speech-to-speech on device 🚀 Updates from Krisp, Yellow.ai, Google, Kudo, Gridspace and much more 🔥

Interesting new project for Speech to Speech pipeline to run GPT4o-like experiences on-device with 500 ms latency 🔥

Top Updates 💪

• AI speech-to-text tools is expected to reach $15B by 2032

• Conversational AI market size, latest trends, emerging technologies, forecast 2030

• FCC proposes new rules for AI-generated content in Calls

• OpenAI’s new voice mode let me talk with my phone, not to it

• Krisp joins NICE CXexchange marketplace

• Yellow.ai launches Analyze to enhance CX automation with deeper insights

• Pindrop launches Pulse inspect: A cutting-edge solution to combat deepfakes

• Google Gemini’s voice chat mode is here

• NICE named a leader in Everest Group’s 2024 Conversational AI PEAK Matrix

• Kudo adds AI speech translation for Microsoft Teams

• Text-to-speech brain implant restores ALS patient's voice

• Google Meet will take notes for you now, thanks to AI

• Gridspace enhances Grace with groundbreaking emotional intelligence

• Hearview revolutionizes communication for the deaf with AI smart glasses.

• Voiceline raises €2.4M in seed funding

• EliseAI raises $75M in Series D funding

• Revoize raises €458K in pre-seed funding
  

Noteworthy 💪

• How to identify AI-generated speech

• ChatGPT’s Advanced Voice Mode. It’s fun, and just a bit creepy

• Emotional context improves translation quality of LLMs

• PH contact centers under pressure from AI

• Another look at iOS 18.1 & Apple intelligence

• How has audio evolved in the last 30 years?

• Sound like a human: The future of voice modulation is here

• AI Chatbot: Bridging the gap between technology and human interaction

• 10 highlights of using PLAUD NOTE AI audio device

  

---

Science and Demo Corner 😎

• Robust speech-to-text, running locally on Quest VR headset

• Speaker diarization using OpenAI Whisper

• Parler: An open source TTS for lightweight applications

• Filmora: Remove background noise from videos

• Whisper-Medusa: aiOla’s STT model, claiming to be 50% faster than Whisper

• Boggl AI: Voice-powered AI assistant for effortless product management