Dexa/Practical AI

Published Dec 4, 2024

Full-duplex, real-time dialogue with Kyutai

Explore the innovation behind Moshi, a groundbreaking real-time speech-to-speech AI assistant from Kyutai with full-duplex dialogue capabilities, as co-founder Alexandre de Fousset delves into the technical challenges and successes. Additionally, discover how QTAI is transforming the AI landscape in France by fostering open-source research and educational growth.

Episode Highlights

Topics covered

Questions from this episode

- What opportunities does AI open up?
  Asked by 7 people
- How does AI transcription work in the context of the episode 1024: Blake Eastman | Can Machines Read People Better Than Humans? and the clip Variable Trauma Insights?
  Asked by 5 people
- Why is openness important for artificial intelligence?
  Asked by 4 people
- Tell me something unique about large language models in the context of the episode 972: Mustafa Suleyman | Navigating The 21st Century's Greatest Dilemma and the clip Language Models' Predictive Power
  Asked by 3 people
- Is this chatbot a large language model (LLM)?
  Asked by 3 people
- What techniques are used with large language models (LLMs)?
  Asked by 2 people
- Which large language model are you based on?
  Asked by 2 people
- Can you give more examples of large language model applications in relation to the episode Stephen Wolfram: ChatGPT and the Nature of Truth, Reality & Computation | Lex Fridman Podcast #376 and the clip Computational Language Insights?
  Asked by 2 people
- What types of AI models are used in podcast AI?
  Asked by 2 people
- Tell me something unique about large language models
  Asked by 2 people
- What trends are shaping AI startups?
  Asked by 2 people
- What are the key differences between AI Chat GPT and other types of AI as discussed in the episode #148 ESAI and the clip AI in Admissions?
  Asked by 1 person
- How does AI transcription work?
  Asked by 1 person
- Why is openness important for artificial intelligence, as discussed in the episode DisTrO and the Quest for Community-Trained AI Models and the clip Open Source AI?
  Asked by 1 person
- How are large language models (LLMs) trained on podcasts?
  Asked by 1 person

Episode Highlights

Related Episodes

The first real-time voice assistant
Answers 383 questions
Real-time conversational insights from phone call data
Answers 383 questions
Towards high-quality (maybe synthetic) datasets
Answers 383 questions
Creating tested, reliable AI applications
Answers 383 questions
Speech recognition to say it just right
Answers 383 questions
Pausing to think about scikit-learn & OpenAI o1
Answers 383 questions
Understanding what's possible, doable & scalable
Answers 383 questions
Gemini vs OpenAI
Answers 383 questions
End-to-end cloud compute for AI/ML
Answers 383 questions
Clones, commerce & campaigns
Answers 383 questions
Autonomous fighter jets?!
Answers 383 questions
Deep-dive into DeepSeek
Answers 383 questions
25 years of speech technology innovation
Answers 383 questions
Going full bore with Graphcore!
Answers 383 questions
The path towards trustworthy AI
Answers 383 questions