Is Attention always needed? A Case Study on Language Identification from Speech

TL;DR

This paper explores the effectiveness of a convolutional recurrent neural network for language identification from speech, comparing it with other models, and demonstrates high accuracy across multiple languages and noisy conditions.

Contribution

It introduces a CRNN-based language identification model and provides a comparative analysis with CNN and attention-based models, showing superior performance and robustness.

Findings

Achieved over 98% accuracy on 13 Indian languages.

High performance (97-100%) on linguistically similar languages.

Robustness to noise with 91.2% accuracy in noisy conditions.

Abstract

Language Identification (LID) is a crucial preliminary process in the field of Automatic Speech Recognition (ASR) that involves the identification of a spoken language from audio samples. Contemporary systems that can process speech in multiple languages require users to expressly designate one or more languages prior to utilization. The LID task assumes a significant role in scenarios where ASR systems are unable to comprehend the spoken language in multilingual settings, leading to unsuccessful speech recognition outcomes. The present study introduces convolutional recurrent neural network (CRNN) based LID, designed to operate on the Mel-frequency Cepstral Coefficient (MFCC) characteristics of audio samples. Furthermore, we replicate certain state-of-the-art methodologies, specifically the Convolutional Neural Network (CNN) and Attention-based Convolutional Recurrent Neural Network (CRNN with attention), and conduct a comparative analysis with our CRNN-based approach. We conducted comprehensive evaluations on thirteen distinct Indian languages and our model resulted in over 98\% classification accuracy. The LID model exhibits high-performance levels ranging from 97% to 100% for languages that are linguistically similar. The proposed LID model exhibits a high degree of extensibility to additional languages and demonstrates a strong resistance to noise, achieving 91.2% accuracy in a noisy setting when applied to a European Language (EU) dataset.