One-against-All Weighted Dynamic Time Warping for Language-Independent and Speaker-Dependent Speech Recognition in Adverse Conditions

Abstract

Considering personal privacy and difficulty of obtaining training material for many seldom used English words and (often non-English) names, language-independent (LI) with lightweight speaker-dependent (SD) automatic speech recognition (ASR) is a promising option to solve the problem. The dynamic time warping (DTW) algorithm is the state-of-the-art algorithm for small foot-print SD ASR applications with limited storage space and small vocabulary, such as voice dialing on mobile devices, menu-driven recognition, and voice control on vehicles and robotics. Even though we have successfully developed two fast and accurate DTW variations for clean speech data, speech recognition for adverse conditions is still a big challenge. In order to improve recognition accuracy in noisy environment and bad recording conditions such as too high or low volume, we introduce a novel one-against-all weighted DTW (OAWDTW). This method defines a one-against-all index (OAI) for each time frame of training data and applies the OAIs to the core DTW process. Given two speech signals, OAWDTW tunes their final alignment score by using OAI in the DTW process. Our method achieves better accuracies than DTW and merge-weighted DTW (MWDTW), as 6.97% relative reduction of error rate (RRER) compared with DTW and 15.91% RRER compared with MWDTW are observed in our extensive experiments on one representative SD dataset of four speakers' recordings. To the best of our knowledge, OAWDTW approach is the first weighted DTW specially designed for speech data in adverse conditions.

Proposed combination of PCA and MFCC feature extraction in speech recognition system

“…Then, the pre-emphasized speech signal is divided into short time segments called "frames" [5]. The length of the overlapping part between adjacent frames, as shown in Fig.…”

“…ICA), Linear Predictive Coding, Cepstral Analysis, Filter bank analysis, kernel-based methods [2] are the common feature extraction methods and have been widely employed in speech recognition research field. Among these methods, the MFCC method has been widely used and has achieved the high recognition accuracy in speech recognition system [3,4,5]. There are many works studying MFCC, especially in improving the recognition accuracy [3].…”

“…2. Please note that the this method can be found in [1,4,5]. In this section, we will give the brief overview of the conventional MFCC feature extraction method.…”

Feature trajectory dynamic time warping for clustering of speech segments

“…In an effort to improve performance, several variations of DTW have been proposed since its inception. For example, a one-against-all index (OAI) for each time series under consideration is proposed in [4]. The OAI is subsequently used to weight the corresponding DTW alignment score in a speech recognition system.…”

A real-time spike classification method based on dynamic time warping for extracellular enteric neural recording with large waveform variability

“…DTW has proven to be very effective in speech recognition under source and ambient variations (Zhang et al (2014)) and ECG profile characterization (Huang and Kinsner (2002)), which are similar to the ENS EAP analysis at hand. We established a fastDTW method with automatic thresholding for ENS recordings with large spike variability for real-time applications.…”