Summary:

• About 1.5 years of experience in research and development of signal processing algorithms for speech/audio/hearing applications
• Excellent understanding of DSP, RF, Digital communications, VLSI design principles and hands on experience with C/MatLab/HDL/Assembly programming
• Working experience with design of speech/audio processing and speech recognition algorithms, for cochlear implants, and hearing aids
• Possess strong problem solving, analytical, debugging, good presentation and organization skills, along with good academic track record and experience

Education:

M.S. EE
B. Tech. E.C.E.

Technical Skills:

Programming Languages: C, Data Structures, Basics of JAVA, Perl
Hardware Descriptive Languages: VHDL, Verilog
Operating Systems: Windows 95/2000/XP/VISTA/WIN7, Mac OSX, MS-DOS, Solaris
Simulators and IDE: MatLab, Simulink, P-Spice, Xilinx, Modelsim, LabVIEW, Cygwin, MS Visual Studio (MSVC)
EDA Tools: Cadence, Virtuso, RF Simulation Tools (AWR, AXIEM)

Relevant Courses:

Digital Signal Processing, Random Processes, Digital Communications, Applied DSP, Speech Signal Processing, Linear Signals and Systems, VLSI Design, Speech and Speaker Recognition, Engineering of Packet Switching Networks, Digital Image Processing, Microwave Engg, Antennas and Wave Propagation, RF Engineering, Radar Engg, Satellite Communications, Digital and Linear IC’s, Embedded Systems.

Master's Thesis: “Impact of selective amplification of consonants on speech intelligibility in noise by hearing impaired listeners”

• Designed and developed several signal processing methods to process the speech consonants to study and improve the speech intelligibility
• Investigated about the influence of selective consonant amplification on speech recognition performance of hearing-impaired listeners
• Conducted experiments to study speech recognition performance in hearing impaired and normal hearing listeners and designed methods to improve

Publications:

Saripella, R., Loizou, P. C., Thibodeau, L. and Alford, J.A. “The effects of selective consonant amplification on sentence recognition in noise by hearing-impaired listeners,” submitted to Journal of the Acoustical Society of America - under review: MS #10-09784.

Work Experience:

2011 – Researcher Confidential, Berkeley, CA, USA.

• Experience with audio and speechsignal modeling for improved pitch/tone perception with normal and impaired hearing
• Verification of auditory nerve models using iterative ripple noise methods
• Design of filter banks & spectral analysis methods for audio/speech perception

Projects:
Designed, implemented and tested the following projects as a part of course work.

• Projects in DSP: Linear filtering using block convolution methods and signal extraction from noise, DTMF transceiver design, FIR filtering using windows, time delay estimation in Radar using Autocorrelation
• Projects in Speech Signal Processing: Monaural Speech Segregation, Estimation of pitchand formants of given speech signal, Speech segregation according to voiced/unvoiced/speech signal using Rabiners algorithm, Implementation of Levinson-Durbin algorithm
• Adaptive Echo Cancellation for Audio: Implemented an adaptive filter in MatLab to remove the echo present in Audio data
• Missing Feature Theory: Implemented and studied missing feature theory for improved robustness in HMM based speech recognition systems
• Signal Predictor: Designed and implemented a Neural-network based signal prediction system using LabVIEW
• Projects in VLSI:
• Basic cell library design, with inverter, NAND, NOR, XOR, XNOR, OAI12, AOI22 and D-Flip Flop and wrote test bench to check their functionality
• Designed of Packet Queuing/Dispatching system & 16-Symbols Huffman Decoder in Verilog and simulated using Modelsim
• Drew schematics and layout using pre-developed library cells, wrote RTL code, checked DRC and LVS, Verified the circuit’s functionality at layout level and behavioral level using Cadence’s Encounter
• Measured power, timing and identified the critical paths using Pathmill
• HMM based Automatic Speech Recognizer: Worked on acoustic speech modeling, mel-frequency cepstral coefficients, HMM modeling, n-word speech recognition for building 8-word automatic speech recognition (ASR) system and evaluated the performance of the system using automated Perl scripts
• ASR system training: Worked on feature extraction, training the ASR systems with bigram/trigram/Ngram language models during speech recognition course
• FFT implementation in VLSI: Designed a Radix-2, 16 Point, Discrete-In-Frequency Fast Fourier Transform in VHDL using XILINX, synthesized and implemented it using Modelsim simulator
• Image Scaling Algorithms: Implemented image down scaling algorithm on YUV420 image based on average area filtering in C using MS visual studio. The algorithm was tested with variousresolutionsof YUV/standard test images
• RF Transceiver Design: Designed a linear amplifier module and patch antenna for 3.0 GHz transceiver application using Lange’s Coupler, Balanced Amplifiers, and Chebyshev filter employing self-biasing using a single power source