Speech Recognition: Statistical Methods

Rabiner, L. R.; Juang, Biing‐Hwang

doi:10.1016/b0-08-044854-2/00907-x

Cited by 24 publications

(17 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this work, the phonetic transcription labels produced by the SRIs Decipher state-of-the-art ASR system [34] 36.1% respectively. While these results are equivalent to those obtained by other state-ofthe-art systems on similar databases [35], transcription errors will be non negligible and will produce that, in order to compute the i-vector for a particular linguistic unit, some frames belonging to a different one will be taken into account, degrading the performance of the system based on that unit. In this work, no exhaustive analysis has been done regarding whether the errors occurred are associated with particular units or speakers, as we have no transcriptions available for the datasets used.…”

Section: Region Conditioningmentioning

confidence: 54%

Linguistically-constrained formant-based i-vectors for automatic speaker recognition

Franco-Pedroso

González-Rodríguez

2016

Speech Communication

View full text Add to dashboard Cite

This is the author’s version of a work that was accepted for publication in Speech Communication. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Speech Communication, VOL 76 (2016) DOI 10.1016/j.specom.2015.11.002This paper presents a large-scale study of the discriminative abilities of formant frequencies for automatic speaker recognition. Exploiting both the static and dynamic information in formant frequencies, we present linguistically-constrained formant-based i-vector systems providing well calibrated likelihood ratios per comparison of the occurrences of the same isolated linguistic units in two given utterances. As a first result, the reported analysis on the discriminative and calibration properties of the different linguistic units provide useful insights, for instance, to forensic phonetic practitioners. Furthermore, it is shown that the set of units which are more discriminative for every speaker vary from speaker to speaker. Secondly, linguistically-constrained systems are combined at score-level through average and logistic regression speaker-independent fusion rules exploiting the different speaker-distinguishing information spread among the different linguistic units. Testing on the English-only trials of the core condition of the NIST 2006 SRE (24,000 voice comparisons of 5 minutes telephone conversations from 517 speakers -219 male and 298 female-), we report equal error rates of 9.57 and 12.89% for male and female speakers respectively, using only formant frequencies as speaker discriminative information. Additionally, when the formant-based system is fused with a cepstral i-vector system, we obtain relative improvements of ∼6% in EER (from 6.54 to 6.13%) and ∼15% in minDCF (from 0.0327 to 0.0279), compared to the cepstral system alone.This work has been supported by the Spanish Ministry of Economy and Competitiveness (project CMC-V2: Caracterizacion, Modelado y Compensacion de Variabilidad en la Señal de Voz, TEC2012-37585-C02-01). Also, the authors would like to thank SRI for providing the Decipher phonetic transcriptions of the NIST 2004, 2005 and 2006 SREs that have allowed to carry out this work

show abstract

Section: Region Conditioningmentioning

confidence: 54%

Linguistically-constrained formant-based i-vectors for automatic speaker recognition

Franco-Pedroso

González-Rodríguez

2016

Speech Communication

View full text Add to dashboard Cite

show abstract

“…The ASR problem is formulated within the Bayesian framework. In this method, an utterance is represented by some sequence of acoustic feature vector X, derived from the underlying sequence of words W, and the recognition system needs to find the most likely word sequence as given below [37]…”

Section: Gmm/dnnmentioning

confidence: 99%

Convolutional Neural Networks for Raw Speech Recognition

Passricha¹,

Aggarwal²

2018

From Natural to Artificial Intelligence - Algorithms and Applications

View full text Add to dashboard Cite

State-of-the-art automatic speech recognition (ASR) systems map the speech signal into its corresponding text. Traditional ASR systems are based on Gaussian mixture model. The emergence of deep learning drastically improved the recognition rate of ASR systems. Such systems are replacing traditional ASR systems. These systems can also be trained in end-to-end manner. End-to-end ASR systems are gaining much popularity due to simplified model-building process and abilities to directly map speech into the text without any predefined alignments. Three major types of end-to-end architectures for ASR are attention-based methods, connectionist temporal classification, and convolutional neural network (CNN)-based direct raw speech model. In this chapter, CNN-based acoustic model for raw speech signal is discussed. It establishes the relation between raw speech signal and phones in a data-driven manner. Relevant features and classifier both are jointly learned from the raw speech. Raw speech is processed by first convolutional layer to learn the feature representation. The output of first convolutional layer, that is, intermediate representation, is more discriminative and further processed by rest convolutional layers. This system uses only few parameters and performs better than traditional cepstral feature-based systems. The performance of the system is evaluated for TIMIT and claimed similar performance as MFCC.

show abstract

“…Actually, there are two possible types of speech feature modelling, deterministic and statistical models (Rabiner and Juang 2006). The deterministic models exploit the intrinsic properties of the speech signal, while the statistical models are concerned with statistical properties of the speech signal (Chien and Chueh 2009).…”

Section: Introductionmentioning

confidence: 99%

Wavelet network for recognition system of Arabic word

Ejbali

Zaied

Amar

2010

Int J Speech Technol

View full text Add to dashboard Cite

Focusing on the development of new technologies of information, research in the speech communication field is an activity in full expansion. Several disciplines and skills interact in order to improve performance of Human Machine Communication Systems (HMC). In order to increase the performance of these systems, various techniques, including Hidden Markov Models (HMM) and Neural Network (NN), are implemented.In this paper, we advance a new approach for modelling of acoustic units and a new method for speech recognition, especially recognition of Arabic word, adapting to this new type of modelling based on Wavelet Network (WN). The new recognition system is a hybrid classifier. It is based on NN as a general model and the wavelets assume the role of activation function.Our approach of speech recognition is divided into two parts: training, and recognition phases. The training stage is based on audio corpus. After converting all training signals from original format to a specific parameterisation, each acoustic vector will be modelled by WN. These vectors will refine and cover all signal properties in one model. It consists in generating a WN for every training signal. The recognition phase is divided into three steps. The first is to extract features from the input vector to be recognized. The second is to estimate all resulting vectors from training WN.The third is to evaluate the distance between the vector to be recognized and the reconstructed vectors.The obtained results shows that our system, based on WN, is very competitive compared to systems based on HMM.

show abstract

Speech Recognition: Statistical Methods

Cited by 24 publications

References 44 publications

Linguistically-constrained formant-based i-vectors for automatic speaker recognition

Linguistically-constrained formant-based i-vectors for automatic speaker recognition

Convolutional Neural Networks for Raw Speech Recognition

Wavelet network for recognition system of Arabic word

Contact Info

Product

Resources

About