Investigation of Speaker-adaptation methods in Transformer based ASR
This work addresses speaker adaptation for end-to-end ASR systems, offering incremental improvements in accuracy for speech recognition tasks.
The paper tackled the problem of improving automatic speech recognition performance by incorporating speaker information into a Transformer-based model, resulting in reduced word error rates on the NPTEL and Librispeech datasets.
End-to-end models are fast replacing the conventional hybrid models in automatic speech recognition. Transformer, a sequence-to-sequence model, based on self-attention popularly used in machine translation tasks, has given promising results when used for automatic speech recognition. This paper explores different ways of incorporating speaker information at the encoder input while training a transformer-based model to improve its speech recognition performance. We present speaker information in the form of speaker embeddings for each of the speakers. We experiment using two types of speaker embeddings: x-vectors and novel s-vectors proposed in our previous work. We report results on two datasets a) NPTEL lecture database and b) Librispeech 500-hour split. NPTEL is an open-source e-learning portal providing lectures from top Indian universities. We obtain improvements in the word error rate over the baseline through our approach of integrating speaker embeddings into the model.