Locale Encoding For Scalable Multilingual Keyword Spotting Models

02/25/2023
by   Pai Zhu, et al.
0

A Multilingual Keyword Spotting (KWS) system detects spokenkeywords over multiple locales. Conventional monolingual KWSapproaches do not scale well to multilingual scenarios because ofhigh development/maintenance costs and lack of resource sharing.To overcome this limit, we propose two locale-conditioned universalmodels with locale feature concatenation and feature-wise linearmodulation (FiLM). We compare these models with two baselinemethods: locale-specific monolingual KWS, and a single universalmodel trained over all data. Experiments over 10 localized languagedatasets show that locale-conditioned models substantially improveaccuracy over baseline methods across all locales in different noiseconditions.FiLMperformed the best, improving on average FRRby 61 similarsizes.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
08/24/2021

Are the Multilingual Models Better? Improving Czech Sentiment with Transformers

In this paper, we aim at improving Czech sentiment with transformer-base...
research
09/08/2022

Multilingual Transformer Language Model for Speech Recognition in Low-resource Languages

It is challenging to train and deploy Transformer LMs for hybrid speech ...
research
02/06/2020

Irony Detection in a Multilingual Context

This paper proposes the first multilingual (French, English and Arabic) ...
research
05/23/2023

When Does Monolingual Data Help Multilingual Translation: The Role of Domain and Model Scale

Multilingual machine translation (MMT), trained on a mixture of parallel...
research
04/08/2020

Structure-Level Knowledge Distillation For Multilingual Sequence Labeling

Multilingual sequence labeling is a task of predicting label sequences u...
research
07/23/2018

ASR-free CNN-DTW keyword spotting using multilingual bottleneck features for almost zero-resource languages

We consider multilingual bottleneck features (BNFs) for nearly zero-reso...
research
11/14/2018

Almost Zero-Resource ASR-free Keyword Spotting using Multilingual Bottleneck Features and Correspondence Autoencoders

We compare features for dynamic time warping based keyword spotting in a...

Please sign up or login with your details

Forgot password? Click here to reset