|
Abstract
|
|
|---|---|
| Language Identification (LID) can be defined as the process for automatically identifying the language of a given spoken utterance. We have focused in a phonotactic approach in which the system input is the phonemes sequence generated by a speech recognizer (ASR), but instead phonemes we have used phonetic units that contain context information ?phone-grams?. In this context, we propose the use of Neural Embeddings (NEs) as features for those phone-grams sequences, which are used as entries in a classical i-Vectors framework to train a multi class logistic classifier. These NEs incorporate information from the neighbouring phone-grams in the sequence and model implicitly longer-context information. The NEs have been trained using both a Skip-Gram and a Glove Model. Experiments have been carried out on the KALAKA-3 database and we have used Cavg as metric to compare the systems. We propose as baseline the Cavg obtained using the NEs as features in the LID task, 24,69%. Our strategy to incorporate information from the neighbouring phone-grams to define the final sequences contributes to obtain up to 24,3% relative improvement over the baseline using Skip-Gram model and up to 32,4% using Glove model. Finally, the fusion of our best system with a MFCC-based acoustic i-Vectors system provides up to 34,1% improvement over the acoustic system alone. | |
|
International
|
Si |
|
Congress
|
IberSpeech 2018 |
|
|
970 |
|
Place
|
Barcelona - España |
|
Reviewers
|
Si |
|
ISBN/ISSN
|
|
|
|
10.21437/IberSPEECH.2018-12 |
|
Start Date
|
21/11/2018 |
|
End Date
|
23/11/2018 |
|
From page
|
55 |
|
To page
|
59 |
|
|
Proceedings IberSPEECH 2018 |