Exploring End-To-End Attention-Based Neural Networks For Native Language Identification

Ubale, Rutuja; Ye, Qian; Evanini, Keelan

doi:10.1109/slt.2018.8639689

Cited by 13 publications

(3 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many studies on language identification have been conducted, with various feature extraction and classification techniques being used. Several techniques are used to extract features from the audio data, including phone recognition followed by language modeling (PRLM) [5] and parallel phone recognition followed by language modeling (PPRLM) [5] for phonetic approach or perceptual linear prediction (PLP) [5], mel-frequency cepstral coefficient (MFCC) [6]- [8], i-vector [8], [9] and x-vector [10] for the acoustic approx neural networks [11], convolutional neural networks (CNN) [12], [13], logistic regression (LR) [8], PLDA [14], Gaussian mixture model (GMM) [15], [16], support vector machine [17], [18] are among techniques used to classify the language spoken.…”

Section: Introductionmentioning

confidence: 99%

Spoken language identification on 4 Indonesian local languages using deep learning

Wijonarko

Zahra

2022

Bulletin EEI

View full text Add to dashboard Cite

Language identification is at the forefront of assistance in many applications, including multilingual speech systems, spoken language translation, multilingual speech recognition, and human-machine interaction via voice. The identification of indonesian local languages using spoken language identification technology has enormous potential to advance tourism potential and digital content in Indonesia. The goal of this study is to identify four Indonesian local languages: Javanese, Sundanese, Minangkabau, and Buginese, utilizing deep learning classification techniques such as artificial neural network (ANN), convolutional neural network (CNN), and long-term short memory (LSTM). The selected extraction feature for audio data extraction employs mel-frequency cepstral coefficient (MFCC). The results showed that the LSTM model had the highest accuracy for each speech duration (3 s, 10 s, and 30 s), followed by the CNN and ANN models.

show abstract

Section: Introductionmentioning

confidence: 99%

Spoken language identification on 4 Indonesian local languages using deep learning

Wijonarko

Zahra

2022

Bulletin EEI

View full text Add to dashboard Cite

show abstract

“…There are several steps to identify language, starting from cleaning the data from noise to help the system get better accuracy, extracting the feature from speech data, and classifying the language. There are several techniques to classify the language spoken, including neural networks [2], convolutional neural networks [3]- [6], logistic regression [7], PLDA [8], gaussian mixture model [9], [10], support vector machine [11], [12], and several techniques to extract the features from the recording, such as MFCC [13], [14], ivector [15]- [17], and x-vector [18].…”

Section: Introductionmentioning

confidence: 99%

Spoken language identification using i-vectors, x-vectors, PLDA and logistic regression

Abdurrahman

Zahra

2021

Bulletin EEI

View full text Add to dashboard Cite

In this paper, i-vector and x-vector is used to extract the features from speech signal from local Indonesia languages, namely Javanese, Sundanese and Minang languages to help classifier identify the language spoken by the speaker. Probabilistic linear discriminant analysis (PLDA) are used as the baseline classifier and logistic regression technique are used because of prior studies showing logistic regression has better performance than PLDA for classifying speech data. Once these features are extracted. The feature is going to be classified using the classifier mentioned before. In the experiment, we tried to segment the test data to three segment such as 3, 10, and 30 seconds. This study is expanded by testing multiple parameters on the i-vector and x-vector method then comparing PLDA and logistic regression performance as its classifier. The x-vector has better score than i-vector for every segmented data while using PLDA as its classifier, except where the i-vector and x-vector is using logistic regression, i-vector still has better accuracy compared to x-vector.

show abstract

“…With the invent of convolution neural networks (CNN) in dialect classification [19], [20] which can handle variable length utterances along with classification, three stages reduced to two. In [19], CNNs are evaluated over the Arabic database (MGB-3) with various acoustic features such asmel-frequency cepstral coefficients (MFCC), log mel-scale filterbank energies (FBANK), and spectrogram.…”

Section: Introductionmentioning

confidence: 99%

Learning Filterbanks from Raw Waveform for Accent Classification

Kethireddy

Kadiri

Gangashetty

2020

2020 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

Most of the applications in speech use mel-frequency spectral coefficients (MFSC) as features as they match the human perceptual mechanism, where the emphasis is given to vocal tract characteristics. But in accent classification, mel-scale distribution of filters may not always be the best representations, e.g., pitch accented languages where the emphasis should be on vocal source information too. Motivated by this, we use end-toend classification of accents directly from waveforms which will reduce the effort of designing features specific to each corpus. The convolution neural network (CNN) model architecture is designed in such a way that the initial layers exhibit similar operation as in MFSC by initializing the weights using time approximate of MFSC. The entire network along with initial layers is trained to learn accent classification. We observed that learning directly from waveform improved the performance of accent classification when compared to CNN trained on hand-engineered features by 10.94% UAR on the test dataset of common voice corpus. Analyzing the filters after learning, we observed changes in distribution and bandwidths of center frequencies. We further observed the importance of appropriately initializing CNN filters.

show abstract

Exploring End-To-End Attention-Based Neural Networks For Native Language Identification

Cited by 13 publications

References 20 publications

Spoken language identification on 4 Indonesian local languages using deep learning

Spoken language identification on 4 Indonesian local languages using deep learning

Spoken language identification using i-vectors, x-vectors, PLDA and logistic regression

Learning Filterbanks from Raw Waveform for Accent Classification

Contact Info

Product

Resources

About