Acoustic Word Embeddings for Low Resource Speech Processing with Herman Kamper - TWiML Talk #191
The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)
00:00
Is There a Difference Between the Zero Resource Speech Recognition Community and Cognitive Science?
There's a lot of evidence that shows that infants can pick up things like syllables even before they understand a complete language. So we know from these lab ments that children are able to figure out longer spanning am word segments. And what we've done in our own work is we've said, while okif if a child can do this am, can we actually build that into our model?
Transcript
Play full episode
