AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
3min chapter
Neural Synthesis of Binaural Speech From Mono Audio with Alexander Richard - #514
The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)
Bringing Convolution Into Neuro Networks
In audio, dynamic time wopping is a very old traditional technique where you want to allin signals in some way. And we essentially incorporate that as a neura network layer as a fully differ able neur network layer to solve this face problem. Because thi time warping allows you to shift components of the audio signal left or right, depending on where they want to have them. We also tried towe talking about noise in both he amplitude direction and the a directionabsolutely. So yes, we kind of shot for a solution that is on the modelling side. But apart from that, it's really this full setting of having mono signal and just the positions and getting a complete bynora
00:00
Transcript
Episode notes
Get the Snipd
podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any
moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share
& Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode