AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
4min chapter
Alexander Veysov on Self-Teaching AI and Creating Open Speech-To-Text
The Gradient: Perspectives on AI
The State of Speech to Text
The main bottleneck is the same, it is data, and obviously, in order to build a system that can be used in production, you have to have real, in-demand data. There have appeared to train models by media, by Facebook, which are not really good. The majority of these systems were published, like as an existing building tools. But I wouldn't say that all of this was designed to bridge the gap between your design and the system. It's more disappointing when people do not understand what effort you have to make after you have downloaded this public model,. And usually people will need that since there was this pretend model, then everything should be free.
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