326 research outputs found
A hybrid CNN-LiGRU acoustic modeling using raw waveform sincnet for Hindi ASR
Deep Neural Network (DNN) is currently playing the most vital role in Automatic Speech Recognition (ASR). Convolution Neural Network (CNN) and Recurrent Neural Network (RNN) are the advanced versions of DNN. CNN and RNN are right to deal with spatial and temporal properties of the speech signal, respectively, and both properties have a higher impact on accuracy. In today’s scenario, many acoustic modeling techniques often switches due to the battle of CNNs and RNNs. In the last few years, CNN, with raw speech signal, shows their superiority over precomputed acoustic features. Recently, a novel first convolution layer named SincNet was proposed to produce the interpretable filters with better accuracy. In this work, we proposed a hybrid SincNet-CNN-RNN architecture with low computation cost and high accuracy. Different configurations of the hybrid model were extensively examined to achieve this goal. All experiments were performed on the Hindi speech dataset
Bi-Lingual TDNN-LSTM Acoustic Modeling for Limited Resource Hindi and Marathi Language ASR
Social sector expenditure in India and Punjab: Trends during pre-reforms (1980-1991) and post-reforms (1991-2011) periods
Integration of Virtual Learning of Induction Machines for Undergraduates
In context of understanding problems faced by undergraduate students while carrying out laboratory experiments dealing with high voltages, it was found that most of the students are hesitant to work directly on machine. The reason is that error in the circuitry might lead to deterioration of machine and laboratory instruments. So, it has become inevitable to include modern pedagogic techniques for undergraduate students, which would help them to first carry out experiment in virtual system and then to work on live circuit. Further advantages include that students can try out their intuitive ideas and perform in virtual environment, hence leading to new research and innovations. In this paper, virtual environment used is of MATLAB/Simulink for three-phase induction machines. The performance analysis of three-phase induction machine is carried out using virtual environment which includes Direct Current (DC) Test, No-Load Test, and Block Rotor Test along with speed torque characteristics for different rotor resistances and input voltage, respectively. Further, this paper carries out computer aided teaching of basic Voltage Source Inverter (VSI) drive circuitry. Hence, this paper gave undergraduates a clearer view of experiments performed on virtual machine (No-Load test, Block Rotor test and DC test, respectively). After successful implementation of basic tests, VSI circuitry is implemented, and related harmonic distortion (THD) and Fast Fourier Transform (FFT) of current and voltage waveform are studied
The antithrombotic properties of polymer-coated, drug eluting coronary stents.
Abstract not available
Recommended from our members
Access to Computer Technologies at Home Improves Wages in the Marketplace
This study suggests that having access to a computer and the Internet at home leads to better wages. We normally have the idea that people with higher wages have computers and Internet access at home. But with declining prices of home PCs and the increasing importance of Internet, this no longer seems to be the case. The new owners of PC today are considerably poorer and less educated than in the past. People with access to a PC and Internet at home appear to earn 8.6% more than their counterpart. This result seems to suggest that University policies requiring students seeking professional degrees like MBA to own a laptop or Home PC has merit and this policy could be extended to cover undergraduate students also. Universities could also provide dial-in-server access to their recent graduates for a specified period of time to boost their placement statistics
Acoustic modeling problem for automatic speech recognition system: advances and refinements (Part II)
- …
