While multilingual BERT outperforms older task-specific models on C-S tasks, its performance on C-S data is significantly worse than on the same tasks in a monolingual setting. This inefficiency occurs because these models are either not exposed to C-S data at all during training, or they miss out on several language pairs.

Inefficiency of Multilingual BERT on C-S Tasks

Since C-S languages tend to be low-resourced, building deep learning-based models is challenging due to the lack of large C-S datasets.

Challenge of Deep Learning for C-S Languages

Massive multilingual language models, such as multilingual BERT, have shown promise in enabling the coverage of low-resource languages without any labeled data, by using a zero-shot framework.

Massive Multilingual Language Models (LMs)

Limitation:  the lack of end-to-end userfacing applications that interact directly with users in multilingual communities.

1) Although computational implementations of grammatical models of C-S exist (Bhat et al., 2016), they do not necessarily generate natural C-S utterances that a bilingual speaker would produce

2) Most crucially, current computational approaches to C-S language technologies do not usually take into account the linguistic and social factors that influence why and when speakers/users choose to code-switch.

User-facing applications

Language technologies cannot yet process code-switching (C-S)—alternating between two or more languages within a single conversation—as seamlessly as monolingual data. The current state of computational C-S processing faces three main limitations: data (scarce C-S datasets), evaluation (models such as multilingual BERT underperform on C-S relative to monolingual settings), and user-facing applications (few end-to-end tools serve multilingual communities).

University of Michigan - Ann Arbor

Google

C-S across Languages: Indian Context

e crucial information about the social and cultural dynamics in a multilingual setting would also be lost

ignoring these studies and data could have serious consequences

perform a quantitative study of 18 recent Bollywood (Hindi) movies and find that C-S is used for establishing identity, social dynamics between characters and the socio-cultural context of the movie

the types and functions of C-S between Hindi and English across the TV programs

Investigations into code-switching (C-S) within Indian written media, advertising, and film reveal a generational shift in language dominance. While Hindi was the primary language with occasional switches to English in early productions, English became the dominant language in later productions, particularly among younger generations.

Evolution of Code-Switching in Indian Media

In the Indian multilingual education system, high-cost English Medium (EM) education is highly valued by affluent families. Code-switching (C-S) is commonly practiced among students in schools across India, making it unrealistic to expect students to rigidly separate their spoken languages.

Learn Before

Related

Learn After