TY  - RPRT
TI  - Improvements to Embedding-Matching Acoustic-to-Word ASR Using Multiple-Hypothesis Pronunciation-Based Embeddings
AU  - Hao Yen
AU  - Woojay Jeon
PY  - 2023
UR  - https://arxiv.org/abs/2210.16726
ID  - 2210.16726
ER  -