Papers by Hemakumar Gopal
— this paper addresses the problem of large vocabulary speaker dependent isolated Kannada words r... more — this paper addresses the problem of large vocabulary speaker dependent isolated Kannada words recognition using the syllables, Hidden Markov Model (HMM) and Normal fit method. This experiment has covered 5.5 million words among the 10 million words from Hampi text corpus. Here 3-state Baum–Welch algorithm is used for training. For the 2 successor outputted λ(A, B, pi) is combined and passed into normal fit, the outputted normal fit parameter is labelled has syllable or sub-word. Our model is compared with Gaussian Mixture Model and HMM (3-state Baum–Welch algorithm). This paper clearly shows that for normal fit applied for HMM will reduce the memory size while building the speech models and works with excellent recognition rate. The average WRR is 91.22 % and average WER is 8.78%. All computations are done using mat lab.
— this paper addresses the problem of Kannada speech signal segmentation. The designed algorithm ... more — this paper addresses the problem of Kannada speech signal segmentation. The designed algorithm automatically segments the continuous Kannada speech signal into syllables and sub-words using the dynamic threshold computation by the combination of short time energy and magnitude of signal. The proposed method first pre-processes the original Kannada speech signal then framing is done for every 20 millisecond with an overlapping of 6.5 millisecond. Secondly frames are passed through hamming window using the same size of frame. Thirdly voiced and unvoiced part is detected through computing dynamic threshold using short time energy and magnitude of signal. In this paper 20 unique sentences are used for experiment which can be used as commands to simple mobile sets. Each of these sentences was recorded for 10 times and segmentation testing is done for each signal and computed accuracy rate of segmentation. The automatic segmentation success rate of individually uttered of sentences in e...
Speaker Accent and Isolated Kannada Word Recognition
Computer Science and Information Technology, 2014
Algorithm is designed for isolated Kannada word recognition of five districts Kannada speakers’ a... more Algorithm is designed for isolated Kannada word recognition of five districts Kannada speakers’ accent. Isolated Kannada words recognition is designed using the syllables, Baum-Welch algorithm and Normal fit method. The novelty of proposed method is in recognition of five district Kannada speaker accents as well as spoken words. Our model is compared with baseline Hidden Markov Model (HMM) and Gaussian Mixture Model (GMM). Our model is tested for both noisy (little) and noiseless signal. For experiment totally 7056 signals are used for training and 3528 signals are used for testing. The experiment showed that average WRR for known accent speaker is 95.56% and of unknown accent speaker is 90.86% and average recognition of Kannada speaker accents is 82%.
Large Vocabulary in Continuous Speech Recognition Using HMM and Normal Fit
International Journal of Computer Trends and Technology, 2016
this paper addresses the problem of large vocabulary speaker independent continuous speech recogn... more this paper addresses the problem of large vocabulary speaker independent continuous speech recognition using the phonemes, Hidden Markov Model (HMM) and Normal fit method. Here we first detect for the voiced part in speech signal through computing dynamic threshold in each frame. Real Cepstrum coefficients are extracted as features from the voiced frames. The Baum–Welch algorithm is applied for training those features. Then normal fit technique is applied, the outputted values are labelled using correspondent phoneme or syllable. The model is tested for 5 languages namely English, Kannada, Hindi, Tamil and Telugu. The automatic segmentation of speech signals average accuracy rate is 95.42% and miss rate of about 4.58%. In the large vocabulary, average Word Recognition Rate (WRR) is 85.16% and average Word Error Rate (WER) is 14.84%. All computations are done using mat lab.
Speaker Independent Isolated Kannada Word Recognizer
Lecture Notes in Electrical Engineering, 2013
This paper addresses the problem of recognizing spoken Kannada words. The designed algorithm reco... more This paper addresses the problem of recognizing spoken Kannada words. The designed algorithm recognizes spoken Kannada words independent of speakers. The proposed method normalizes the original speech signal of every isolated word and extracts Linear-Predictive coding (LPC) coefficients, and converts them into Real Cepstrum Coefficient. These Real Cepstrum Coefficient values are subjected to dimensionality reduction through normal fit. These coefficients are used as the representatives of each spoken word. Euclidian distance measure is then used to compute the distance between the test samples to the model data in the database. The model datum in the database at a minimum distance is declared as the recognized word. For experimentation, we have used 294 unique Kannada words. Each of these words was recorded with 10 Speakers yielding 2,940 samples in total. Out of 10 speakers’ data, 8 speakers’ data i.e., 2,352 samples were used to compute the representative co-efficient for each word. Remaining 2 speakers’ data along with re-recorded data of two speakers out of the 8 speakers is used for testing. Totally 2,352 signals are used for training and 1,176 signals are used for testing. The success rate of the proposed system- known speaker data is 98.29 % and unknown speaker data is 91.66 %.
ijcsi.org
this paper discusses a speech recognition based on spoken English words formed by vowel, diphthon... more this paper discusses a speech recognition based on spoken English words formed by vowel, diphthong and plosive and it has been developed and experimented for single speaker. The success rate of recognition of individually uttered words in experiments is excellent and has reached about 98.86 %. The miss rate of about 1.14% was almost only because of false acceptance. In phonemes classification on an average we have reached 85% and miss classification rate was 15%. We have successfully tested all the words formed by vowel followed by vowel-plosive or plosive-vowels or diphthongplosive and reached high success rate in recognition of the words. All computations are performed in MATLAB and PRAAT software.
This paper made a study on academic leadership and found that leadership in academic field poses ... more This paper made a study on academic leadership and found that leadership in academic field poses problems that are noticeably different than leadership in government agencies and companies.Paper absorbed that academic leaders need to work closely with teaching, learning, research, and scholarship to bring out the best among academics. A literature survey is done on academic leadership in higher education (HE) and issues on freedom of academic, which shows the great importance and relevance in this context. The paper discusses the University Governance Architecture and New Education Policy (NEP) for leadership and effective governance in HEIs, Academic Leadership, and selection of administrators according to NEP-2020. Paper discussed Transforming Regulatory System of HE in NEP-2020, Financing HEIs through the new body called HEGC (Higher Education Grants Council), General Education Council, PSSB (Professional Standard-Setting Bodies) designed in NEP-2020, and the Institutional Develo...
Acoustic Phonetic Characteristics of Kannada Language
ijcsi.org
... 4. Vowels alone can become meaning word, but consonants alone cannot become words orakshara. ... more ... 4. Vowels alone can become meaning word, but consonants alone cannot become words orakshara. A consonant needs help of vowels to become akshara. 5. The Liquid Trill phone /r/ is the only one phone which works with any phoneme combination. ... A sp irated ...
During the period of corona, technology has made Indian education to find considerable changes, i... more During the period of corona, technology has made Indian education to find considerable changes, it has modified the educational system from traditional chalk, and board based teaching to online education system. The old method of class rooms teaching with black boards are replaced by laptops and smart phones, which are now connected with advanced audio and visual system presentation. Teachers are doing online teaching with the assistance of teach mint, google classroom, google meet, zoom, etc., which are providing classroom feeling to students. Virtual teaching and learning skills are changing the direction of education system. Digitization of education system has produced both advantages and disadvantages in dynamic environment of our educational sector. However, here attempt has been made to find the challenges to be faced by both teachers and students. This paper considers only secondary data and information has been gathered from various journals, websites, newspapers and magazines.
NOESIS INTERNATIONAL SCIENTIFIC RESEARCH JOURNAL, 2021
This paper has discussed about the growth of new educational policies and their implementation's ... more This paper has discussed about the growth of new educational policies and their implementation's across the nation. Paper delivers some statistics regarding several students, teachers, and educational institutions across the Nation, which shows the importance of the New Education Policy (NEP-2020). It also engrossed some major problems in previous policy which is urgently needed to be overcome for the betterment of the young mindset's to contest with growing competition in the global knowledge and economy. Hence, the 2020 NEP 1 lays particular importance on the progress of the innovative potential of each individual. The NEP is based on the idea that education should focus not only on cognitive abilities, but also on foundational capacities' like knowledge and proficiency, as well as higher-order intellectual abilities like critical thinking and problem solving, as well as social, ethical, and emotional aptitudes and temperaments. This paper has given more importance regarding the reforms on examinations, due to every learning students need to go for assessment and evaluated using the exams only. As the primary method of education, we rely on routine memorization. Schools continue to use test scores as the major criterion for assessing students' understanding and expertise. As a result, the educational system as a whole is relatively dictatorial. Everyone believe that good test scores are the key to achieving success. Two examples used to explain the teaching-learning-evaluation process held and how the qualities of examination should be, and evaluating different personalities in the same contest is needed or not. Then discuss, the changes required in the examinations process and focusing on multilingualism in schools education as well as exams. The difference between the assessment and evaluation, assessment and evaluation reforms, tools applied for evaluation. The transforming the culture of assessment, performance describes, formative assessments. NEP reforming examinations in Grades 9 to 12, cognitive domain testing, psychomotor domain testing, affective domaininvolves feelings and attitudes. Finally, paper has suggested the requirement of planning and designing the new progress card in a new fashion which has to be recorded and reflected in the 360degree multi-dimension report.
This paper made a study on academic leadership and found that leadership in academic field poses ... more This paper made a study on academic leadership and found that leadership in academic field poses problems that are noticeably different than leadership in government agencies and companies.Paper absorbed that academic leaders need to work closely with teaching, learning, research, and scholarship to bring out the best among academics. A literature survey is done on academic leadership in higher education (HE) and issues on freedom of academic, which shows the great importance and relevance in this context.
Uploads
Papers by Hemakumar Gopal