Using Adaptive Filter to Increase Automatic Speech Recognition Rate in a Digit Corpus

Many new consumer applications are based on the use of automatic speech recognition (ASR) systems, such as voice command interfaces, speech-to-text applications, and data entry processes. Although ASR systems have remarkably improved in recent decades, the speech recognition system performance still significantly degrades in the presence of noisy environments. Developing a robust ASR system that can work in real-world noise and other acoustic distorting conditions is an attractive research topic. Many advanced algorithms have been developed in the literature to deal with this problem; most of these algorithms are based on modeling the behavior of the human auditory system with perceived noisy speech. In this research, the power-normalized cepstral coefficient (PNCC) system is modified to increase robustness against the different types of environmental noises, where a new technique based on gammatone channel filtering combined with channel bias minimization is used to suppress the noise effects. The TIDIGITS database is utilized to evaluate the performance of the proposed system in comparison to the state-of-the-art techniques in the presence of additive white Gaussian noise (AWGN) and seven different types of environmental noises. In this research, one word is recognized from a set containing 11 possibilities only. The experimental results showed that the proposed method provides significant improvements in the recognition accuracy at low signal to noise ratios (SNR). In the case of subway noise at SNR = 5 dB, the proposed method outperforms the mel-frequency cepstral coefficient (MFCC) and relative spectral (RASTA)–perceptual linear predictive (PLP) methods by 55% and 47%, respectively. Moreover, the recognition rate of the proposed method is higher than the gammatone frequency cepstral coefficient (GFCC) and PNCC methods in the case of car noise. It is enhanced by 40% in comparison to the GFCC method at SNR 0dB, while it is improved by 20% in comparison to the PNCC method at SNR −5dB.

Download Full-text

Automatic speech signal segmentation based on the innovation adaptive filter

International Journal of Applied Mathematics and Computer Science ◽

10.2478/amcs-2014-0019 ◽

2014 ◽

Vol 24 (2) ◽

pp. 259-270 ◽

Cited By ~ 9

Author(s):

Ryszard Makowski ◽

Robert Hossa

Keyword(s):

Speech Recognition ◽

Automatic Speech Recognition ◽

Adaptive Filter ◽

Speech Signal ◽

Detection Efficiency ◽

Difficult Problem ◽

Second Order ◽

Signal Segmentation ◽

Schur Algorithm ◽

Starting Point

Abstract Speech segmentation is an essential stage in designing automatic speech recognition systems and one can ﬁnd several algorithms proposed in the literature. It is a difﬁcult problem, as speech is immensely variable. The aim of the authors’ studies was to design an algorithm that could be employed at the stage of automatic speech recognition. This would make it possible to avoid some problems related to speech signal parametrization. Posing the problem in such a way requires the algorithm to be capable of working in real time. The only such algorithm was proposed by Tyagi et al., (2006), and it is a modiﬁed version of Brandt’s algorithm. The article presents a new algorithm for unsupervised automatic speech signal segmentation. It performs segmentation without access to information about the phonetic content of the utterances, relying exclusively on second-order statistics of a speech signal. The starting point for the proposed method is time-varying Schur coefﬁcients of an innovation adaptive ﬁlter. The Schur algorithm is known to be fast, precise, stable and capable of rapidly tracking changes in second order signal statistics. A transfer from one phoneme to another in the speech signal always indicates a change in signal statistics caused by vocal track changes. In order to allow for the properties of human hearing, detection of inter-phoneme boundaries is performed based on statistics deﬁned on the mel spectrum determined from the reﬂection coefﬁcients. The paper presents the structure of the algorithm, deﬁnes its properties, lists parameter values, describes detection efﬁciency results, and compares them with those for another algorithm. The obtained segmentation results, are satisfactory.

Download Full-text

Usage of Prosody Modification and Acoustic Adaptation for Robust Automatic Speech Recognition (ASR) System

Revue d intelligence artificielle ◽

10.18280/ria.350307 ◽

2021 ◽

Vol 35 (3) ◽

pp. 235-242

Author(s):

Vivek Bhardwaj ◽

Vinay Kukreja ◽

Amitoj Singh

Keyword(s):

Speech Recognition ◽

Automatic Speech Recognition ◽

Research Work ◽

Recognition Rate ◽

Speaking Rate ◽

Word Error Rate ◽

Adaptive Training ◽

Prosody Modification ◽

Children's Speech ◽

Asr System

Most of the automatic speech recognition (ASR) systems are trained using adult speech due to the less availability of the children's speech dataset. The speech recognition rate of such systems is very less when tested using the children's speech, due to the presence of the inter-speaker acoustic variabilities between the adults and children's speech. These inter-speaker acoustic variabilities are mainly because of the higher pitch and lower speaking rate of the children. Thus, the main objective of the research work is to increase the speech recognition rate of the Punjabi-ASR system by reducing these inter-speaker acoustic variabilities with the help of prosody modification and speaker adaptive training. The pitch period and duration (speaking rate) of the speech signal can be altered with prosody modification without influencing the naturalness, message of the signal and helps to overcome the acoustic variations present in the adult's and children's speech. The developed Punjabi-ASR system is trained with the help of adult speech and prosody-modified adult speech. This prosody modified speech overcomes the massive need for children's speech for training the ASR system and improves the recognition rate. Results show that prosody modification and speaker adaptive training helps to minimize the word error rate (WER) of the Punjabi-ASR system to 8.79% when tested using children's speech.

Download Full-text

Indonesian Automatic Speech Recognition For Command Speech Controller Multimedia Player

EMITTER International Journal of Engineering Technology ◽

10.24003/emitter.v2i2.25 ◽

2014 ◽

Vol 2 (2) ◽

Author(s):

Vivien Arief Wardhany ◽

Sritrusta Sukaridhoto ◽

Amang Sudarsono

Keyword(s):

Speech Recognition ◽

Automatic Speech Recognition ◽

English Language ◽

Language Model ◽

Recognition Rate ◽

Male Voice ◽

Similar Word ◽

The Difference ◽

Model Recognition ◽

Speech Recognizer

The purpose of multimedia devices development is controlling through voice. Nowdays voice that can be recognized only in English. To overcome the issue, then recognition using Indonesian language model and accousticc model and dictionary. Automatic Speech Recognizier is build using engine CMU Sphinx with modified english language to Indonesian Language database and XBMC used as the multimedia player. The experiment is using 10 volunteers testing items based on 7 commands. The volunteers is classifiedd by the genders, 5 Male & 5 female. 10 samples is taken in each command, continue with each volunteer perform 10 testing command. Each volunteer also have to try all 7 command that already provided. Based on percentage clarification table, the word “Kanan” had the most recognize with percentage 83% while “pilih” is the lowest one. The word which had the most wrong clarification is “kembali” with percentagee 67%, while the word “kanan” is the lowest one. From the result of Recognition Rate by male there are several command such as “Kembali”, “Utama”, “Atas “ and “Bawah” has the low Recognition Rate. Especially for “kembali” cannot be recognized as the command in the female voices but in male voice that command has 4% of RR this is because the command doesn’t have similar word in english near to “kembali” so the system unrecognize the command. Also for the command “Pilih” using the female voice has 80% of RR but for the male voice has only 4% of RR. This problem is mostly because of the different voice characteristic between adult male and female which male has lower voice frequencies (from 85 to 180 Hz) than woman (165 to 255 Hz).The result of the experiment showed that each man had different number of recognition rate caused by the difference tone, pronunciation, and speed of speech. For further work needs to be done in order to improving the accouracy of the Indonesian Automatic Speech Recognition system.Keywords: Automatic Speech Recognizer, Indonesian Acoustic Model, CMU Sphinx, indonesian Language Model, Recognition Rate, XBMC.

Download Full-text