Abstract

Language identification (LID) systems, which can model highlevel information such as phonotactics have exhibited superior performance. State-of-the-art models use sequential models to capture the high-level information, but these models are sensitive to the length of the utterance and do not equally generalize over variable length utterances. To effectively capture this information, a feature that can model the long-term temporal context is required. This study aims to capture the long-term temporal context by appending successive shifted delta cepstral (SDC) features. Deep neural networks have been explored for developing LID systems. Experiments have been performed using AP17-OLR database. LID systems developed by stacking SDC features have shown significant improvement compared to the system trained with SDC features. The proposed feature with residual connections in the feed-forward networks reduced the equal error rate from 21.04, 18.02, 16.45 to 14.42, 11.14 and 10.11 on the 1-second, 3-seconds and > 3-second test utterances respectively.

Abstract

This work focuses on the development of speech data comprising 23 Indian languages for developing language identification (LID) systems. Large data is a pre-requisite for developing state-of-the-art LID systems. With this motivation, the task of developing multilingual speech corpus for Indian languages has been initiated. This paper describes the composition of the data and the performances of various LID systems developed using this data. In this paper, Mel frequency cepstral feature representation is used for language identification. In this work, various state-of-the-art LID systems are developed using i-vectors, deep neural network (DNN) and deep neural network with attention (DNN-WA) models. The performance of the LID system is observed in terms of the equal error rate for i-vector, DNN and DNN-WA is 17.77%, 17.95%, and 15.18% respectively. Deep neural network with attention model shows a better performance over i-vector and DNN models

Abstract

In this study, the acoustic-phonetic attributes of palatalization in the Kashmiri speech is investigated. It is a unique phonetic feature of Kashmiri in the Indian context. An automated approach is proposed to detect this unique phonetic feature from the continuous Kashmiri speech. The i-matra vowel has the impact of palatalizing the consonant connected to it. Therefore, these consonants investigated in synchronous with vowel regions, which are spotted using the instantaneous energy computed from the envelope-derivative of the speech signal. The resonating characteristics of the vocal-tract system framework that reflect the formant dynamics are used to differentiate palatalized consonants from the other consonants. In this regard, the Hilbert envelope of the numerator of the group-delay function that provides good time-frequency resolution used to extract formants. The palatalization detection experimentation carried out in various vowel contexts using the acoustic cues, and it produced a promising result with a detection accuracy of 92.46 %.

Abstract

Automatic language identification (LID) in practical environments is gaining a lot of scientific attention due to rapid developments in multilingual speech processing applications. When an LID is operated in noisy environments a degradation in the performance can be observed and it can be majorly attributed to mismatch between the training and operating environments. This work is aimed towards developing an LID system that can robustly operate in clean and noisy environments. Traditionally, to reduce the mismatch between training and operating environments, noise is synthetically induced to the training corpus and these models are termed as multi-SNR models. In this work, various curriculum learning strategies are explored to train multi-SNR models, such that the trained models have better generalization in performance over varying background environments. I-vector, Deep neural networks (DNN) and DNN With Attention (DNN-WA) architectures are used in this work for developing LID systems. Experimental verification of the proposed approach is carried out using IIIT-H Indian database and AP17-OLR database. The performance of LID system is tested at different signal-to-noise ratio (SNR) levels using white and vehicular noises from NOISEX dataset. In comparison to multi-SNR models, the LID systems trained with curriculum learning have performed better in terms of equal error rate (EER) and generalization in EER across varying background environments. The degradation in the performance of LID systems due to environmental noise has been effectively reduced by training multi-SNR models using curriculum learning.

Abstract

Vowels are produced with an open configuration of the vocal tract, without any audible friction. The acoustic signal is relatively loud with varying strength of impulselike excitation. Vowels possess significant energy content in the low-frequency bands of the speech signal. Acoustic events such as vowel onset point (VOP) and vowel end-point (VEP) can be used as landmarks to detect vowel regions in a speech signal. In this paper, two-stage algorithm is proposed to detect precise vowel regions. In the first level, the speech signal is processed using zero frequency filtering to emphasize energy content in low-frequency bands of speech. Zero frequency filtered signal predominantly contains lowfrequency content of the speech signal as it is filtered around 0 Hz. This process is followed by the extraction of dominant spectral peaks from the magnitude spectrum around glottal closure regions of the speech signal. The vowel onset points and vowel end-points are obtained by convolving the enhanced spectral contour of zero frequency filtered signal with first order Gaussian differentiator. In the next level, a post-processing is carried out in the regions around VOP and VEP to remove spurious vowel regions based on uniformity of epoch intervals. In addition, the positions of VOPs and VEPs are also corrected using the strength of the excitation of the speech signal. The performance of the proposed vowel region detection method is compared with the existing state of art methods on TIMIT acoustic-phonetic speech corpus. It is reported that this method produced significant improvement in vowel region detection in clean and noisy environments.

Abstract

The physiological constraints in the human speech production mechanism restrict the emotional state of the speaker for a short duration in an emotional speech. In this work, the emotionally stressed regions are detected using the strength of the excitation (SoE) and fundamental frequency (F0) of the speech signal. These emotionally stressed regions for the different emotions are further classified using a linear kernel support vector machine (SVM) with a twostage binary decision logic. This classifier is modelled using the differenced prosody features extracted by considering the relative difference of the prosody components between normal and emotionally stressed regions. This experimentation is carried out using Berlin emotion speech (EMO-DB) database. The proposed approach produced a better result compared to existing state-of-the-art emotion recognition systems. An accuracy of 80.83% is reported in this paper.

Abstract

In this study, a robust language identification (LID) system is investigated with vowel region detection scheme incorporated at the front-end. The background noise is known to severely reduce language identification performance. In this work, Mel-frequency cepstral features extracted from the vowel regions instead of voiced speech are utilized for language identification system modeling. Experimentation is done using different vowel region detection algorithms on deep neural networks (DNN) and deep neural networks with attention (DNN-WA) based LID systems on IIITH Indian language dataset. Experimental results show that the accurate vowel region detection mechanism at the front-end of a LID system improved the performance in noisy environment.

Abstract

In this paper, a novel technique has been proposed for the vowel region detection from the continuous speech using an envelope of the derivative of the speech signal, which is a non-negative, frequency-weighted energy operator. The proposed vowel region detection method is implemented using a two-stage algorithm. The first stage of vowel region detection consists of speech signal analysis to detect vowel onset points (VOP) and vowel end-points (VEP) using an instantaneous energy contour obtained from the envelope of the derivative of a speech signal. The VOPs and VEPs are spotted using the peak-finding algorithm based upon the first order Gaussian differentiator. The next stage consists of removal of spurious vowel regions and the correction of hypothesized VOP and VEP locations using combined cues obtained from the uniformity of epoch intervals and strength of the excitation of the speech signal. Performance of the proposed method for detecting vowel regions from the speech signal is evaluated using TIMIT acoustic-phonetic speech corpus. The proposed approach resulted in significantly high detection rate and less false alarm rate compared to the state-of-the-art methods in both clean and noisy environments

Abstract

t Phonetic feature extraction is used in many types of speech applications. In this paper, we propose an approach for automatic detection of retroflex approximant /õ/ in a Tamil continuous speech using sonorant regions and formant structure of speech. Retroflex approximant is one of the distinctive phonetic features of Tamil speech. It occurs in V_V and V_ contexts. In the proposed approach, slopes of the formant trajectories and spectral dynamics of F2 and F3 in the region of retroflex approximant are considered as acoustic cues for detecting them in continuous speech. In this work, numerator of the group delay function-based formant extraction method is used to track formant trajectories. Based on experimental results, formants extracted using this approach produced better evidence compared to state-of-the-art formant extraction techniques for retroflex approximant detection.

Abstract

Speaker normalization is one of the crucial aspects of an Automatic speech recognition system (ASR). Speaker normalization is employed to reduce the performance drop in ASR due to speaker variabilities. Traditional speaker normalization methods are mostly linear transforms over the input data estimated per speaker, such transforms would be efficient with sufficient data. In practical scenarios, only a single utterance from the test speaker is accessible. The present study explores speaker normalization methods for end-to-end speech recognition systems that could efficiently be performed even when single utterance from the unseen speaker is available. In this work, it is hypothesized that by suitably providing information about the speaker’s identity while training an end-to-end neural network, the capability to normalize the speaker variability could be incorporated into an ASR system. The efficiency of these normalization methods depends on the representation used for unseen speakers. In this work, the identity of the training speaker is represented in two different ways viz. i) by using a one-hot speaker code, ii) a weighted combination of all the training speakers identities. The unseen speakers from the test set are represented using a weighted combination of training speakers representations. Both the approaches have reduced the word error rate (WER) by 0.6, 1.3% WSJ corpus