# Phone-aware Neural Language Identification

## TL;DR

This paper introduces a phone-aware neural language identification system that leverages phonetic information from an ASR system to significantly improve language detection accuracy, even for unseen languages.

## Contribution

The paper proposes a novel deep LSTM-RNN LID architecture that incorporates phonetic knowledge from an ASR system, enhancing performance over purely acoustic models.

## Key findings

- Significant performance improvement with phonetic knowledge
- Effective even when test language is unseen in ASR training
- Validated on four languages within Babel corpus

## Abstract

Pure acoustic neural models, particularly the LSTM-RNN model, have shown great potential in language identification (LID). However, the phonetic information has been largely overlooked by most of existing neural LID models, although this information has been used in the conventional phonetic LID systems with a great success. We present a phone-aware neural LID architecture, which is a deep LSTM-RNN LID system but accepts output from an RNN-based ASR system. By utilizing the phonetic knowledge, the LID performance can be significantly improved. Interestingly, even if the test language is not involved in the ASR training, the phonetic knowledge still presents a large contribution. Our experiments conducted on four languages within the Babel corpus demonstrated that the phone-aware approach is highly effective.