Detail publikačního výsledku

End-to-End Open Vocabulary Keyword Search With Multilingual Neural Representations

YUSUF, B.; ČERNOCKÝ, J.; SARAÇLAR, M.

Originální název

End-to-End Open Vocabulary Keyword Search With Multilingual Neural Representations

Anglický název

End-to-End Open Vocabulary Keyword Search With Multilingual Neural Representations

Druh

Článek WoS

Originální abstrakt

Conventional keyword search systems operate on automatic speech recognition (ASR) outputs, which causes them to have a complex indexing and search pipeline. This has led to interest in ASR-free approaches to simplify the search procedure. We recently proposed a neural ASR-free keyword search model which achieves competitive performance while maintaining an efficient and simplified pipeline, where queries and documents are encoded with a pair of recurrent neural network encoders and the encodings are combined with a dot-product. In this article, we extend this work with multilingual pretraining and detailed analysis of the model. Our experiments show that the proposed multilingual training significantly improves the model performance and that despite not matching a strong ASR-based conventional keyword search system for short queries and queries comprising in-vocabulary words, the proposed model outperforms the ASR-based system for long queries and queries that do not appear in the training data.

Anglický abstrakt

Conventional keyword search systems operate on automatic speech recognition (ASR) outputs, which causes them to have a complex indexing and search pipeline. This has led to interest in ASR-free approaches to simplify the search procedure. We recently proposed a neural ASR-free keyword search model which achieves competitive performance while maintaining an efficient and simplified pipeline, where queries and documents are encoded with a pair of recurrent neural network encoders and the encodings are combined with a dot-product. In this article, we extend this work with multilingual pretraining and detailed analysis of the model. Our experiments show that the proposed multilingual training significantly improves the model performance and that despite not matching a strong ASR-based conventional keyword search system for short queries and queries comprising in-vocabulary words, the proposed model outperforms the ASR-based system for long queries and queries that do not appear in the training data.

Klíčová slova

Keyword search, spoken term detection, end-to-end keyword search, asr-free keyword search, keyword spotting.

Klíčová slova v angličtině

Keyword search, spoken term detection, end-to-end keyword search, asr-free keyword search, keyword spotting.

Autoři

YUSUF, B.; ČERNOCKÝ, J.; SARAÇLAR, M.

Rok RIV

2024

Vydáno

02.08.2023

Nakladatel

IEEE

Místo

PISCATAWAY, NJ

ISSN

2329-9290

Periodikum

IEEE-ACM Transactions on Audio Speech and Language Processing

Svazek

31

Číslo

08

Stát

Spojené státy americké

Strany od

3070

Strany do

3080

Strany počet

11

URL

https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10201906

BibTex

@article{BUT185202,
  author="YUSUF, B. and ČERNOCKÝ, J. and SARAÇLAR, M.",
  title="End-to-End Open Vocabulary Keyword Search With Multilingual Neural Representations",
  journal="IEEE-ACM Transactions on Audio Speech and Language Processing",
  year="2023",
  volume="31",
  number="08",
  pages="3070--3080",
  doi="10.1109/TASLP.2023.3301239",
  issn="2329-9290",
  url="https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10201906"
}

Dokumenty

yusuf_atm_transactions_2023_vol-31