Data-selective Transfer Learning for Multi-Domain Speech Recognition

09/08/2015
by   Mortaza Doulaty, et al.
0

Negative transfer in training of acoustic models for automatic speech recognition has been reported in several contexts such as domain change or speaker characteristics. This paper proposes a novel technique to overcome negative transfer by efficient selection of speech data for acoustic model training. Here data is chosen on relevance for a specific target. A submodular function based on likelihood ratios is used to determine how acoustically similar each training utterance is to a target test set. The approach is evaluated on a wide-domain data set, covering speech from radio and TV broadcasts, telephone conversations, meetings, lectures and read speech. Experiments demonstrate that the proposed technique both finds relevant data and limits negative transfer. Results on a 6--hour test set show a relative improvement of 4 and 2

READ FULL TEXT
research
07/25/2022

Unsupervised data selection for Speech Recognition with contrastive loss ratios

This paper proposes an unsupervised data selection method by using a sub...
research
09/08/2015

Unsupervised Domain Discovery using Latent Dirichlet Allocation for Acoustic Modelling in Speech Recognition

Speech recognition systems are often highly domain dependent, a fact wid...
research
07/09/2019

Transfer Learning from Audio-Visual Grounding to Speech Recognition

Transfer learning aims to reduce the amount of data required to excel at...
research
07/02/2019

Latent Dirichlet Allocation Based Acoustic Data Selection for Automatic Speech Recognition

Selecting in-domain data from a large pool of diverse and out-of-domain ...
research
09/04/2019

VoipLoc: Establishing VoIP call provenance using acoustic side-channels

We develop a novel technique to determine call provenance in anonymous V...
research
02/28/2023

Exploring Self-supervised Pre-trained ASR Models For Dysarthric and Elderly Speech Recognition

Automatic recognition of disordered and elderly speech remains a highly ...
research
04/01/2021

Interactive spatial speech recognition maps based on simulated speech recognition experiments

In their everyday life, the speech recognition performance of human list...

Please sign up or login with your details

Forgot password? Click here to reset