Moshi beats OpenAI's voice model 👀 Updates from NVIDIA, Twilio, MS, ElevenLabs and much more 🔥

Moshi’s AI voice assistant beat OpenAI to one of ChatGPT's most anticipated features. According to Kyutai, Moshi can speak in various accents and has 70 different emotional and speaking styles.

Top Updates 💪

• Conversational AI market to grow $87B by 2034

  

• Speech-to-text API market estimated to reach $12.1 billion by 2031

• NVIDIA NeMo T5-TTS model tackles hallucinations in speech synthesis

• Twilio announces a mobile app for CCaaS, “Personalized IVR”, & more

• ElevenLabs Voice Isolator: Remove background noise for film, podcast, interview

• Microsoft to add name mispronunciation detector to Teams

• Philips Speech and Sembly AI join to introduce AI meeting transcription solution

  

• The best AI Meeting Assistant: 12 options for productivity

• Microsoft expands Azure AI Speech with multilingual voices and avatars

  

• Vida, provider of AI voice agents, raises $3M funding

• Best Text-to-Speech software in 2024

Noteworthy 💪

• Zoom vs Zoom Workplace: Is Zoom Workplace Just Zoom?

  

• Are voice notes the latest weapon in the deepfake arsenal?

  

• Get more from audio data with Conversational Intelligence

• Make your voice chatbots more engaging with new text to speech features

• Assessment of Pepper Robot’s STT system through the lens of machine learning

---

Science and Demo Corner 😎

• VALL-E 2: Breakthrough in human-like TTS technology

  

• Moshi: A GPT-4o-like model that can see hear and talk natively

• Enhancing AI-powered video to text transcription on Google Cloud

• Audio AI: The best AI voice generators & text to voice tools

• Central Kurdish TTS with novel end-to-end transformer training

• Enhancing multilingual speech recognition in air traffic control

• Best video translators to translate video 2024

  

• Best Ringly.IO alternatives

• Best Voiceplug.ai alternatives