Jun 26, 2026

Builders Unscripted: Ep. 4 - Pietro Schirano

Pietro Schirano, Founder & CEO of MagicPath, discusses his pioneering work with GPT-5.5 and Codex, transforming creative ideas into software and hardware solutions. He details using advanced AI for image-to-sound conversion, multi-agent workflows, resurrecting obsolete tech with new functionalities, and building his company, MagicPath, on the principle of humans directing AI agents. This interview provides deep insights into the creative potential and practical applications of cutting-edge AI for developers and entrepreneurs.

Jun 09, 2026

From Transcription to Live Music: Gemini's Audio Stack — Thor Schaeff, Google DeepMind

Thor Schaeff from Google DeepMind demos the advanced audio AI stack, starting with a single API call to Gemini for rich transcription (speaker names, emotions, translation). He showcases speech generation directed by "director's notes" instead of a voice catalog, the real-time, sound-to-sound Gemini 1.5 Flash Live model, and a live demo of Gemini Live using the Lyria 2 model as a tool to generate a full song on stage.

Mar 30, 2026

Mistral: Voxtral TTS, Forge, Leanstral, & Mistral 4 — w/ Pavan Kumar Reddy & Guillaume Lample

Mistral's Pavan (Voxtral lead) and Guillaume (Chief Scientist) discuss the new Voxtral TTS model, its novel architecture using flow matching for efficient, high-quality speech generation. They elaborate on Mistral's strategy of delivering specialized, open-weight models and the Mistral Forge platform, which empowers enterprises to leverage their proprietary data through fine-tuning for privacy, cost-effectiveness, and superior performance. The conversation also covers Mistral Small, the future of AI for science, and the company's commitment to open-source and foundational research, including formal proving as a proxy for long-horizon reasoning.

Sep 30, 2025

Introducing Sora 2

A detailed overview of OpenAI's announcement of Sora 2, a flagship video and audio generation model, and the new Sora app, which introduces novel features like "Cameo" for personalized content creation and a new social experience.

Jul 28, 2025

Make some noise: Teaching the language of audio to an LLM using sound tokens

Shivam Mehta from KTH presents a method for teaching Large Language Models (LLMs) to understand and generate audio by treating it as a discrete language. The approach involves a two-step process: first, creating an ultra-low bitrate (0.293 kbps) audio representation using a causal variational autoencoder, and second, fine-tuning a Llama 7B model with these audio tokens using LoRA.