Advancements in AI Speech Models and Transcription Efficiency

In this episode, we sit down with Joey Conway to explore NVIDIA's open source AI, from the reasoning-focused Nemotron models built on top of Llama, to the blazing-fast Parakeet speech model. We chat about what makes open foundation models so valuable, how enterprises can think about deploying multi-model strategies, and why reasoning is becoming the key differentiator in real-world AI applications.

Featuring:

• Joey Conway – LinkedIn
• Chris Benson – Website, LinkedIn, Bluesky, GitHub, X

Links:

• Llama Nemotron Ultra
• NVIDIA Llama Nemotron Ultra Open Model Delivers Groundbreaking Reasoning Accuracy
• Independent analysis of AI
• Parakeet Model
• Parakeet Leaderboard
• Try the Llama-3.1-Nemotron-Ultra-253B-v1 model here and here