Dynamic out-of-vocabulary word registration to language model for speech recognition
| Autor: | Bohan Chen, Norihide Kitaoka, Yuya Obashi |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
Out-of-vocabulary words
Acoustics and Ultrasonics Computer science Speech recognition lcsh:QC221-246 02 engineering and technology 01 natural sciences Out of vocabulary Language model lcsh:QA75.5-76.95 0103 physical sciences lcsh:Acoustics. Sound 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing OOV registration lcsh:Electronic computers. Computer science Electrical and Electronic Engineering 010301 acoustics Word (computer architecture) |
| Zdroj: | EURASIP Journal on Audio, Speech, and Music Processing, Vol 2021, Iss 1, Pp 1-8 (2021) |
| ISSN: | 1687-4722 |
| Popis: | We propose a method of dynamically registering out-of-vocabulary (OOV) words by assigning the pronunciations of these words to pre-inserted OOV tokens, editing the pronunciations of the tokens. To do this, we add OOV tokens to an additional, partial copy of our corpus, either randomly or to part-of-speech (POS) tags in the selected utterances, when training the language model (LM) for speech recognition. This results in an LM containing OOV tokens, to which we can assign pronunciations. We also investigate the impact of acoustic complexity and the “natural” occurrence frequency of OOV words on the recognition of registered OOV words. The proposed OOV word registration method is evaluated using two modern automatic speech recognition (ASR) systems, Julius and Kaldi, using DNN-HMM acoustic models and N-gram language models (plus an additional evaluation using RNN re-scoring with Kaldi). Our experimental results show that when using the proposed OOV registration method, modern ASR systems can recognize OOV words without re-training the language model, that the acoustic complexity of OOV words affects OOV recognition, and that differences between the “natural” and the assigned occurrence frequencies of OOV words have little impact on the final recognition results. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink