Speech Recognition of Isolated Malayalam Words Using Wavelet Features and Artificial Neural Network

Krishnan, V.R.V.; Jayakumar, Athulya; Anto, Puji

doi:10.1109/delta.2008.88

Cited by 21 publications

(9 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recording is done in normal office environment using a head set, having microphone with 70Hz to 1600Hz of frequency range. Moreover, it is done with 16 kHz sampling frequency quantized by 16 bit, using a tool named Praat [16]. The speech is saved in Microsoft wave format…”

Section: Speech Corpusmentioning

confidence: 99%

“…The trigram based language model with back-off is used for recognition. The language model is created using the CMU statistical LM toolkit [16].…”

Section: Language Modelingmentioning

confidence: 99%

“…Recently, there has been a large increase in the number of recognition applications for use over telephones, including automated dialing, operator assistance, and remote data access services; such as financial services, for voice dictation systems like medical transcription applications. Such [16] and number recognition systems [6] are the reported works in Malayalam ASR is a branch of Artificial Intelligence (AI) and is related with number of fields of knowledge such as acoustics, linguistics, pattern recognition etc [12]. Speech is the most complex signal to deal with.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Automated Transcription System for Malayalam Language

Kurian¹,

Balakrishnan²

2011

IJCA

View full text Add to dashboard Cite

Malayalam is one of the 22 scheduled languages in India with more than 130 million speakers. This paper presents a report on the development of a speaker independent, continuous transcription system for Malayalam. The system employs Hidden Markov Model (HMM) for acoustic modeling and Mel Frequency Cepstral Coefficient (MFCC) for feature extraction. It is trained with 21 male and female speakers in the age group ranging from 20 to 40 years. The system obtained a word recognition accuracy of 87.4% and a sentence recognition accuracy of 84%, when tested with a set of continuous speech data.

show abstract

Section: Speech Corpusmentioning

confidence: 99%

“…The trigram based language model with back-off is used for recognition. The language model is created using the CMU statistical LM toolkit [16].…”

Section: Language Modelingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automated Transcription System for Malayalam Language

Kurian¹,

Balakrishnan²

2011

IJCA

View full text Add to dashboard Cite

show abstract

“…A comparison of different spectral analysis models for SR using neural networks was performed by Zebulun et al [18]. Krishnan et al [16] used an ANN for recognition of Malayalam words. Dede and Sazli [19] worked on isolated SR with different ANN topologies.…”

Section: Introductionmentioning

confidence: 99%

“…A number of features have been reported in the literature for SR. The linear predictive cepstral coefficient (LPCC) [15], mel frequency cepstral coefficient (MFCC) [15], and wavelet packet-based coefficients [16] are widely used features.…”

Section: Introductionmentioning

confidence: 99%

Speech recognition using ANN and predator-influenced civilized swarm optimization algorithm

Mittal¹,

Sharma²

2016

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

This paper proposes a hybrid optimization technique, predator-influenced civilized swarm optimization, by integrating civilized swarm optimization (CSO) and predator-prey optimization (PPO) techniques. CSO is the integration of the attributes of particle swarm optimization and a society civilization algorithm (SCA). In the SCA, the swarm is divided into a few societies, and each society has its own society leader (SL); other individuals of the society are termed society members. The combination of all such societies forms a civilization, and the best-performing SL becomes the civilization leader (CL). In CSO, SLs and members update their positions through the guidance of their own CL and SLs, respectively, along with their best positions. In the proposed technique, the PPO technique is integrated with CSO, in which a predator particle is included in the swarm. The predator always tries to chase the CL in a controlled manner, which maintains diversity in the population and avoids local optimum solutions. The proposed optimization technique is applied to optimize the weights and biases of an artificial neural network (ANN) trained for speech recognition. Two databases have been used; one is a TI-46 isolated word database in clean and noisy conditions, and the other is a self-recorded Hindi numeral database. To evaluate the performance of the proposed technique, 2 performance criteria, correlation coefficient and mean square error, are applied. The results obtained by an ANN with the proposed technique outperform the results obtained by an ANN trained with particle swarm optimization, PPO, CSO, and backpropagation techniques in terms of correlation coefficient and mean square error.

show abstract

Information Security for Automatic Speaker Identification

El‐Samie

2011

SpringerBriefs in Speech Technology

View full text Add to dashboard Cite

Speech Recognition of Isolated Malayalam Words Using Wavelet Features and Artificial Neural Network

Cited by 21 publications

References 2 publications

Automated Transcription System for Malayalam Language

Automated Transcription System for Malayalam Language

Speech recognition using ANN and predator-influenced civilized swarm optimization algorithm

Information Security for Automatic Speaker Identification

Contact Info

Product

Resources

About