In this blog post, I will strive to provide a down-to-earth introduction to “exotic” floating-point formats, with a special focus on how they can be leveraged to efficiently train large-scale AI models. By the end of this blog post, you will be aware of both the positive and negative implications... [Read More]

In this blog post, I’ll take you through two important concepts behind modern Audio AI models such as Google’s AudioLM and VALL-E, Meta’s AudioGen and MusicGen, Microsoft’s NaturalSpeech 2, Suno’s Bark, Kyutai’s Moshi and Hibiki, and many more: Neural Audio Codecs and (Residual) Vector Quantization. [Read More]