Adaptable Multi-Domain Language Model for Transformer ASR

Abstract

We propose an adapter based multi-domain Transformer based language model (LM) for Transformer ASR. The model consists of a big size common LM and small size adapters. The model can perform multi-domain adaptation with only the small size adapters and its related layers. The proposed model can reuse the full finetuned LM which is fine-tuned using all layers of an original model. The proposed LM can be expanded to new domains by adding about 2% of parameters for a first domain and 13% parameters for after second domain. The proposed model is also effective in reducing the model maintenance cost because it is possible to omit the costly and time-consuming common LM pre-training process. Using proposed adapter based approach, we observed that a general LM with adapter can outperform a dedicated music domain LM in terms of word error rate (WER).

Findings of the Shared Task on Speech Recognition for Vulnerable Individuals in Tamil

“…Similar work was performed for other languages like Bengali, Japanese, etc. Also, more speech corpus is collected from the young people for many languages (Zeng et al, 2020;Lee et al, 2021). However, speaker fluctuation, environmental noise, and transmission channel noise all degrade ASR performance.…”

SUH_ASR@LT-EDI-ACL2022: Transformer based Approach for Speech Recognition for Vulnerable Individuals in Tamil

“…Additionally, 13% WER is reduced by LSTM decoder (Zeng et al, 2021). Transformer model encoding and decoding can be carried with self-attention and multi-head attention layer (Lee et al, 2021). For CTC/Attention based End-To-End ASR, the transformer model is used, which result 23.66% of WER (Miao et al, 2020).…”