An algorithm is presented for the estimation of the fundamental frequency (F0) of speech or musical sounds. It is based on the well-known autocorrelation method with a number of modifications that combine to prevent errors. The algorithm has several desirable features. Error rates are about three times lower than the best competing methods, as evaluated over a database of speech recorded together with a laryngograph signal. There is no upper limit on the frequency search range, so the algorithm is suited for high-pitched voices and music. The algorithm is relatively simple and may be implemented efficiently and with low latency, and it involves few parameters that must be tuned. It is based on a signal model (periodic signal) that may be extended in several ways to handle various forms of aperiodicity that occur in particular applications. Finally, interesting parallels may be drawn with models of auditory processing.

/pdf/yin-a-fundamental-frequency-estimator-for-speech-and-music-53voam3nec.pdf

YIN, a fundamental frequency estimator for speech and music

In the development process of noise-reduction algorithms, an objective machine-driven intelligibility measure which shows high correlation with speech intelligibility is of great interest. Besides reducing time and costs compared to real listening experiments, an objective intelligibility measure could also help provide answers on how to improve the intelligibility of noisy unprocessed speech. In this paper, a short-time objective intelligibility measure (STOI) is presented, which shows high correlation with the intelligibility of noisy and time-frequency weighted noisy speech (e.g., resulting from noise reduction) of three different listening experiments. In general, STOI showed better correlation with speech intelligibility compared to five other reference objective intelligibility models. In contrast to other conventional intelligibility models which tend to rely on global statistics across entire sentences, STOI is based on shorter time segments (386 ms). Experiments indeed show that it is beneficial to take segment lengths of this order into account. In addition, a free Matlab implementation is provided.

An Algorithm for Intelligibility Prediction of Time–Frequency Weighted Noisy Speech

福祉関連職におけるMaslach Burnout Inventoryの因子構造の比較

(NOTE: Each chapter begins with an introduction and concludes with a Summary, Exercises and Bibliography.) 1. Introduction. Discrete-Time Speech Signal Processing. The Speech Communication Pathway. Analysis/Synthesis Based on Speech Production and Perception. Applications. Outline of Book. 2. A Discrete-Time Signal Processing Framework. Discrete-Time Signals. Discrete-Time Systems. Discrete-Time Fourier Transform. Uncertainty Principle. z-Transform. LTI Systems in the Frequency Domain. Properties of LTI Systems. Time-Varying Systems. Discrete-Fourier Transform. Conversion of Continuous Signals and Systems to Discrete Time. 3. Production and Classification of Speech Sounds. Anatomy and Physiology of Speech Production. Spectrographic Analysis of Speech. Categorization of Speech Sounds. Prosody: The Melody of Speech. Speech Perception. 4. Acoustics of Speech Production. Physics of Sound. Uniform Tube Model. A Discrete-Time Model Based on Tube Concatenation. Vocal Fold/Vocal Tract Interaction. 5. Analysis and Synthesis of Pole-Zero Speech Models. Time-Dependent Processing. All-Pole Modeling of Deterministic Signals. Linear Prediction Analysis of Stochastic Speech Sounds. Criterion of "Goodness". Synthesis Based on All-Pole Modeling. Pole-Zero Estimation. Decomposition of the Glottal Flow Derivative. Appendix 5.A: Properties of Stochastic Processes. Random Processes. Ensemble Averages. Stationary Random Process. Time Averages. Power Density Spectrum. Appendix 5.B: Derivation of the Lattice Filter in Linear Prediction Analysis. 6. Homomorphic Signal Processing. Concept. Homomorphic Systems for Convolution. Complex Cepstrum of Speech-Like Sequences. Spectral Root Homomorphic Filtering. Short-Time Homomorphic Analysis of Periodic Sequences. Short-Time Speech Analysis. Analysis/Synthesis Structures. Contrasting Linear Prediction and Homomorphic Filtering. 7. Short-Time Fourier Transform Analysis and Synthesis. Short-Time Analysis. Short-Time Synthesis. Short-Time Fourier Transform Magnitude. Signal Estimation from the Modified STFT or STFTM. Time-Scale Modification and Enhancement of Speech. Appendix 7.A: FBS Method with Multiplicative Modification. 8. Filter-Bank Analysis/Synthesis. Revisiting the FBS Method. Phase Vocoder. Phase Coherence in the Phase Vocoder. Constant-Q Analysis/Synthesis. Auditory Modeling. 9. Sinusoidal Analysis/Synthesis. Sinusoidal Speech Model. Estimation of Sinewave Parameters. Synthesis. Source/Filter Phase Model. Additive Deterministic-Stochastic Model. Appendix 9.A: Derivation of the Sinewave Model. Appendix 9.B: Derivation of Optimal Cubic Phase Parameters. 10. Frequency-Domain Pitch Estimation. A Correlation-Based Pitch Estimator. Pitch Estimation Based on a "Comb Filter<170. Pitch Estimation Based on a Harmonic Sinewave Model. Glottal Pulse Onset Estimation. Multi-Band Pitch and Voicing Estimation. 11. Nonlinear Measurement and Modeling Techniques. The STFT and Wavelet Transform Revisited. Bilinear Time-Frequency Distributions. Aeroacoustic Flow in the Vocal Tract. Instantaneous Teager Energy Operator. 12. Speech Coding. Statistical Models of Speech. Scaler Quantization. Vector Quantization (VQ). Frequency-Domain Coding. Model-Based Coding. LPC Residual Coding. 13. Speech Enhancement. Introduction. Preliminaries. Wiener Filtering. Model-Based Processing. Enhancement Based on Auditory Masking. Appendix 13.A: Stochastic-Theoretic parameter Estimation. 14. Speaker Recognition. Introduction. Spectral Features for Speaker Recognition. Speaker Recognition Algorithms. Non-Spectral Features in Speaker Recognition. Signal Enhancement for the Mismatched Condition. Speaker Recognition from Coded Speech. Appendix 14.A: Expectation-Maximization (EM) Estimation. Glossary.Speech Signal Processing.Units.Databases.Index.About the Author.

Discrete-Time Speech Signal Processing: Principles and Practice

Phase patterns of neuronal responses reliably discriminate speech in human auditory cortex.

Detection of a signal in the presence of diotic masking noise is much better when the signal differs in terms of its interaural parameters from the masking noise. Besides contributing to the detectability of a signal, interaural time delays (ITDs) also allow the listener to lateralize the signal. In this study, we investigate how detection and lateralization of time-delayed signals are affected by the diotic masking noise. All stimuli covered a spectral range from 250 to 850 Hz. The interaurally time-delayed signals consisted either of noise, or a tone-complex with 20 Hz intercomponent spacing. ITDs were 0.1 or 0.6 msec. Results show that detection performance is strongly dependent on the ITD with a 10 dB increase in threshold going from 0.6 to 0.1 msec ITD. Interestingly, lateralization thresholds, although being higher than detection thresholds, are much less dependent on ITD; only a few dB change in threshold is observed. An additional ITD adjustment experiment was performed where the ITD of a tone complex in isolation had to be adjusted to an ITD tone-complex presented in diotic and roving-ITD masking noise. Listeners were able to perform this adjustment reliably suggesting that in the noise, the tone complex effectively functions as a segregated object.

/pdf/a-comparison-of-detection-and-lateralization-of-interaurally-34ic2jm3pj.pdf

A comparison of detection and lateralization of interaurally time-delayed signals in the presence of diotic masking noise

In this contribution, we investigate the internal representations given by the binaural model proposed by Breebaart et al. [1]. We focus on the cues used by an arti??cial observer when determining just noticeable differences in interaural time differences (ITDs) or interaural level differences (ILDs). The binaural processor consists of an array of excitation-inhibition (EI) cells, each characterized by its internal interaural delay and internal interaural attenuation. This model is conceptually similar to a cross-correlation model and the relevant properties for ITD and ILD discrimination tasks are changes in the binaural pattern along the internal delay and the internal attenuation axis. For ITD discrimination of two coherent signals, the traditional approaches consist of searching for a displacement of the peak of the normalized cross-correlation function or measuring the difference in the cross-correlation function at lag zero. These approaches will be compared to our strategy which maximizes the differences between the two normalized cross-correlation functions. Threshold predictions depend on which position along the internal delay line is analyzed. The characteristics of the binaural pattern with regard to ILD discrimination tasks will also be investigated and compared to those that are relevant for ITD discrimination.

A closer look at the representation of interaural differences in a binaural model

/pdf/the-effect-of-cochlear-nonlinearities-on-binaural-masking-punrr8h1n9.pdf

The effect of cochlear nonlinearities on binaural masking level differences

The research reported here explored a possible influence of interaural differences of a broadband target sound of several different durations on detection of its onset in a diotic broadband masker sound. For two positions in the lateral plane, five target/marker duration combinations and various signal-to-noise ratios, the subject’s ability to accurately align a lateralized Gaussian noise target’s onset to the meter of a regular series of marker pulses in a Gaussian noise masker was investigated. It was found that decreasing the signalto-noise ratio between target and masker yields a coherent degradation in onset positioning accuracy, used as a quantitative measure for detection of the onset, for all the lateralization conditions and target/marker duration combinations applied.

On the influence of interaural differences on onset detection in auditory object formation

The lateralization of tone complexes presented against a diotic band-pass noise background was measured as a function of level and Interaural Time Difference (ITD) of the tone complex The bandwidths of the sine-phase tone complex and the noise were 600 Hz, centered at 550 Hz, the component spacing of the tones was 20 Hz Due to this spectral configuration, none of the tonal components is spectrally resolved, and the binaural cues of the combined stimulus vary with the 20 Hz cycle of the stimulus Results show that listeners can judge the lateral position of the tone complex down to levels of −5 dB SNR A second experiment determined masked thresholds of the tone complex in the diotic noise as a function of ITD Detection thresholds are as low as −20 dB SNR for ITDs of 07 ms, and about −7 dB for ITDs of 0 ms, a condition in which only monaural information could be used by the listeners It seems that despite the audibility of the tone complex at rather low levels (−20 dB SNR) based on binaural cues, the ability to lateralize requires the audibility of monaural envelope information, which is restricted to levels above about −7 dB SNR Thus, monaural envelope information may be a prerequisite for the readout of lateralization information corresponding to the tone complex from the binaural display

AG Armin Kohlrausch

Papers

A comparison of detection and lateralization of interaurally time-delayed signals in the presence of diotic masking noise

A closer look at the representation of interaural differences in a binaural model

The effect of cochlear nonlinearities on binaural masking level differences

On the influence of interaural differences on onset detection in auditory object formation

Lateralization of Tone Complexes in Noise: The Role of Monaural Envelope Processing in Binaural Hearing