Paper: | SLP-P13.2 |
Session: | New Features, Models and Representations / Audio Visual ASR |
Location: | Poster Area A, Ground Floor |
Session Time: | Thursday, May 16, 13:00 - 15:00 |
Presentation Time: | Thursday, May 16, 13:00 - 15:00 |
Presentation: |
Poster
|
Topic: |
Speech Processing: General Topics in Speech Recognition |
Paper Title: |
TRULY UNSUPERVISED ACOUSTIC WORD EMBEDDINGS USING WEAK TOP-DOWN CONSTRAINTS IN ENCODER-DECODER MODELS |
Manuscript Link: |
Click here to view manuscript on IEEE Xplore |
Authors: |
Herman Kamper, Stellenbosch University, South Africa |