Effective Sentence Scoring Method Using BERT for Speech Recognition
By Joo-Ah Shin, Yoonhyung Lee, Kyomin Jung. 2019
113 citations recorded by the source corpus.
Subfields: Natural Language Processing, Speech & Audio
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Natural Language Processing
- Julian Salazar
- Shinsuke Sakai
- Tatsuya Kawahara
- Katrin Kirchhoff
- Hayato Futami
- Masato Mimura
- Davis Liang
- Toan Q. Nguyen
In Speech & Audio
- Toan Q. Nguyen
- Julian Salazar
- Shinsuke Sakai
- Tatsuya Kawahara
- Katrin Kirchhoff
- Masato Mimura
- Davis Liang
- Hirofumi Inaguma
Citations in the corpus
- Masked Language Model Scoring (2019, methodology): “Our work extends the closest previous works (Wang and Cho, 2019; Shin et al., 2019) with regards to experiments and tasks, as outlined in Section 2.1.”
- Distilling the Knowledge of BERT for Sequence-to-Sequence ASR (2020, background): “N -best rescoring with BERT was proposed in [14, 15], but the recognition result was restricted to hypotheses from left-to-right decoding.”
- Pre-Training Transformers as Energy-Based Cloze Models (2020, unknown): “While historically log-likelihoods from language models have been used for such re-ranking, recent work has demonstrated that PLLs from masked language models perform better (Shin et al., 2019).”