Advancing Acoustic-to-Word CTC Model With Attention and Mixed-Units
The acoustic-to-word model based on the Connectionist Temporal Classification (CTC) criterion is a natural end-to-end (E2E) system directly targeting…
Read moreTASLP Volume 27 Issue 12
<p>TASLP Volume 27 Issue 12</p>
Beyond Error Propagation: Language Branching Also Affects the Accuracy of Sequence Generation
Sequence generation tasks, such as neural machine translation (NMT) and abstractive summarization, usually suffer from exposure bias as well as the…
Read moreThree-Dimensional Sound Field Reproduction Based on Weighted Mode-Matching Method
A sound field reproduction method based on the spherical wavefunction expansion of sound fields is proposed, which can be flexibly applied to various…
Read more
