A weighted overlap-add method of short-time Fourier analysis/Synthesis

Crochiere, R.

doi:10.1109/tassp.1980.1163353

Cited by 251 publications

(150 citation statements)

References 13 publications

Supporting

Mentioning

129

Contrasting

Unclassified

Order By: Relevance

“…Phase vocoding and related techniques rely on manipulations of the STFT magnitude. For many of these modifications, however, a valid inverse of the STFT does not exist [Hayes et al, 1980] -resulting in the proliferation of heuristic methods for maintaining phase coherence in the synthesis phase [Crochiere, 1980]. In addition to issues associated with synthesizing signals from filter bank or STFT representations, these approaches can be unwieldy due to the large number of parameters required to model the vocalization.…”

Section: Stimulus Design For Auditory Neuroethologymentioning

confidence: 99%

Efficient Encoding Of Vocalizations In The Auditory Midbrain

Holmstrom¹

2000

View full text Add to dashboard Cite

An important question in sensory neuroscience is what coding strategies and mechanisms are used by the brain to detect and discriminate among behaviorally relevant stimuli. To address the noisy response properties of individual neurons, sensory systems often utilize broadly tuned neurons with overlapping receptive fields at the system's periphery, resulting in homogeneous responses among neighboring populations of neurons. It has been hypothesized that progressive response heterogeneity in ascending sensory pathways is evidence of an efficient encoding strategy that minimizes the redundancy of the peripheral neural code and maximizes information throughput for higher level processing. This hypothesis has been partly supported by the documentation of neural heterogeneity in various cortical structures.This dissertation examines whether selective and sensitive responses to behaviorally relevant stimuli contribute to a heterogeneous and efficient encoding in the auditory midbrain. Prior to this study, no compelling experimental framework ex- My parents, for their support throughout all of my education. Not only did they instill in me a confidence that I could take on any educational challenge, but they supported me when I chose not to take these challenges on. I'm sure they worried about me when I dropped out of college in 1993, but their trust in my own decision making process was a better lesson for me to learn at the time.My sister, for teaching me how to fight for the important things in life.My entire committee, for challenging me to push my limits (which I discovered on many occasions). I have leaned on each and every one of them at different times through this process. I am very grateful to have had access to a group so diverse in their specialties, yet so consistent in their kindness, approachability, depth of knowledge, and eagerness to share this knowledge. I couldn't have done this without a single one of them.Lonneke Eeuwes, for all of the long hours she spent collecting data for our experiii iments and carefully reading through my manuscripts.Sunghan Kim, for working with me through the technical details of our state space frequency tracking methods. My research was made possible by your work.Roberto Santiago, for taking me under his wing as a fresh graduate student and imbuing me with a sense of adventure about research. He taught me that the challenges inherent in science are not a chore, but a privilege to be relished.John Vasallo, for setting time aside for our weekly musical catharsis and for being flexible with respect to my constantly shifting schedule. He almost single handedly ensured that playing music will continue to be an important part of my life.Todd Haynes, for his roving intellect and dedication to the creative process. I can only hope that I can tap into a similar wellspring of enthusiasm, kindness, and vision.

show abstract

Section: Stimulus Design For Auditory Neuroethologymentioning

confidence: 99%

Efficient Encoding Of Vocalizations In The Auditory Midbrain

Holmstrom¹

2000

View full text Add to dashboard Cite

show abstract

“…The SOI possible DOA region is set as [3 ,5 ] to center the nominal SOI DOA ( 4 ) with a span 1 to include the actual DOA but to exclude interference DOAs. All algorithms are implemented in frequency domain block by block with the FFT length 1024, and the time domain output waveform is obtained using the overlap and add method [7]. And every point is measured after convergence of the ANC [4].…”

Section: Simulationmentioning

confidence: 99%

Robust Generalized Sidelobe Canceller with an Eigenanalysis-Based Blocking Matrix

WANG¹,

REN²

2017

dtetr

View full text Add to dashboard Cite

Abstract. In acoustic environments with strong background noise and interferences, the generalized sidelobe canceller (GSC) suffers from signal distortion and reduced amount of noise suppression since the blocking matrix cannot produce pure noise reference due to unavoidable direction of arrival (DOA) mismatch. In this paper, a generalized sidelobe canceller robust to DOA mismatch is proposed. It employs eigenanalysis to reconstruct the interference-plus-noise covariance matrix with the possible DOA region of signal of interest (SOI). The reconstructed matrix is provided for the MaxSINR beamforming and subsequent blocking matrix construction. The designed blocking matrix can minimize the signal leakage meanwhile maximize the noise and interference power for further noise and interference suppression. Numerical simulation results show that compared with the alternatives, the proposed GSC has better interference suppression capability and lower SOI distortion in situations without precise DOA of SOI.

show abstract

“…Modified DFT values are obtained by combining the original DFT magnitudes and modified phase values. Finally IFFT is performed and discrete time signal is obtained using overlap add method [7]. The experiments were conducted using white noise, train noise and babble noise.…”

Section: Experimentation Proceduresmentioning

confidence: 99%

Speech Enhancement in Hearing Aids Using Conjugate Symmetry of DFT and SNR-Perception Models

Apte¹,

Shridhar²

2010

IJCA

View full text Add to dashboard Cite

Most of the speech enhancement algorithms use the magnitude of STFT while phase is kept unchanged [3]. In this paper the magnitude of STFT of noisy speech is kept unchanged while the phase is modified. Modified complex spectrum of speech is obtained by combining unchanged magnitude spectrum and modified phase spectrum [3]. This modification results into cancellation of low energy components (noise) of complex spectrum more than the high-energy (speech) components leading to reduction in background noise which is essential in hearing aids. SNR-Perception models for AWGN, Train noise, and Babble noise are developed using Nonlinear Regression modeling technique.Normal hearing and hearing loss subjective listening tests and spectrogram analysis show that the proposed method gives improved speech quality.

show abstract

A weighted overlap-add method of short-time Fourier analysis/Synthesis

Cited by 251 publications

References 13 publications

Efficient Encoding Of Vocalizations In The Auditory Midbrain

Efficient Encoding Of Vocalizations In The Auditory Midbrain

Robust Generalized Sidelobe Canceller with an Eigenanalysis-Based Blocking Matrix

Speech Enhancement in Hearing Aids Using Conjugate Symmetry of DFT and SNR-Perception Models

Contact Info

Product

Resources

About