Sign in

Multi-level self-attentive TDNN: A general and efficient approach to summarize speech into discriminative utterance-level representations.

João MonteiroJahangir AlamTiago H. Falk
Published in: Speech Commun. (2022)
Keyphrases
  • speech recognition
  • special case
  • higher level
  • spoken language
  • semi supervised
  • data sets
  • computer vision
  • natural language
  • hidden markov models
  • cost effective
  • visual attention
  • noisy environments