Formant Tracking Linear Prediction Model using HMMs for Noisy Speech Processing

Formant tracking linear prediction model using HMMs and Kalman filters for noisy speech processing

In this paper, we propose a robust speech enhancement algorithm for nonstationary noise environments, which comprises a multiband spectral subtraction (MBSS) speech estimator and a minima controlled recursive averaging (MCRA) noise estimation. A multiband approach is obtained by dividing the whole spectrum into multiple subbands and applying spectral subtraction independently in each band. The noise estimate exploit the observation that the noise signal typically has a nonuniform effect on the spectrum of speech and is given by averaging the past spectrum power value and using a time and frequency dependent smoothing factor that is calculated based on the speech presence probability in subbands. We test the performance of proposed algorithm in various nonstationary noise environments. The experiment results confirm the superiority of the MBSS and MCRA estimator. Outstanding speech enhancement is achieved, while avoiding the residual musical noise phenomena.

In this paper, we propose a new method for the accurate estimation and tracking of formants in speech signals using time-varying quasi-closed-phase (TVQCP) analysis. Conventional formant tracking methods typically adopt a two-stage estimateand-track strategy wherein an initial set of formant candidates are estimated using short-time analysis (e.g., 10-50 ms), followed by a tracking stage based on dynamic programming or a linear state-space model. One of the main disadvantages of these approaches is that the tracking stage, however good it may he, cannot improve upon the formant estimation accuracy of the first stage. The proposed TVQCP method provides a single-stage formant tracking that combines the estimation and tracking stages into one. TVQCP analysis combines three approaches to improve formant estimation and tracking: (1) it uses temporally weighted quasi-closed-phase analysis to derive closed-phase estimates of the vocal tract with reduced interference from the excitation source, (2) it increases the residual sparsity by using the L-1 optimization and (3) it uses time-varying linear prediction analysis over long time windows (e.g., 100-200 ms) to impose a continuity constraint on the vocal tract model and hence on the formant trajectories. Formant tracking experiments with a wide variety of synthetic and natural speech signals show that the proposed TVQCP method performs better than conventional and popular formant tracking tools, such as Wavesurfer and Praat (based on dynamic programming), the KARMA algorithm (based on Kalman filtering), and DeepFormants (based on deep neural networks trained in a supervised manner). Matlab scripts for the proposed method can be found at: https://github.com/njaygowda/ftrack

The spectral subtraction method is a conventional approach for single channel speech enhancement. The basic principle of this method is to estimate the short-time spectral magnitude of speech by subtracting estimated noise from the noisy speech spectrum and to combines it with the phase of the noisy speech. Besides reducing the noise, this method generates an unnatural and unpleasant noise, called remnant noise. This paper proposes a novel algorithm to reduce the remnant noise, and thus improving the overall quality of the enhanced speech. In this algorithm, the output of multi-band spectral subtraction (MBSS) method is used as the input signal again for next iteration process. After the MBSS method, the additive noise is changed to remnant noise. The remnant noise is re-estimated at each iteration. The new estimated noise, furthermore, is been used to process the next MBSS. This procedure is iterated a small number of times. The simulation results as well as informal subjective evaluations prove that the speech enhanced by proposed algorithm is more pleasant to listeners than the conventional MBSS algorithm. This reveals that the proposed algorithm reduces remnant noise satisfactorily and produces good speech quality with improved signal-to-noise ratio.

Improving the line-spectrum detection capability of a single hydrophone is of great significance for the passive detection of small underwater platforms. In this paper, we propose a single-hydrophone cross-power spectrum (SHCS) method based on time-domain coherence. This method uses the coherence of the line spectrum and the non-coherence of the continuous spectrum noise to obtain coherent gain and improve the signal-to-noise ratio (SNR) of the line spectrum. The effects of the input SNR, number of averaging operations, and overlap ratio on the performance of the SHCS method under a background of Gaussian white noise are simulated and analyzed. The results show that when the overlap ratio is 0 and the number of averaging operations reaches saturation, the SHCS method can achieve the best performance and about 15 dB coherence gain is obtained. The performance of the SHCS method was verified by sea experiments. Under the extremely low input SNR, in which the line spectrum was almost completely submerged in the marine environmental noise, the SHCS method can obtain about 10 dB coherence gain. Under the conventional input SNR, in which the line spectrum could be observed, the SHCS method can obtain about 13 dB coherence gain. The results of processing the radiated noise from an actual cargo ship also demonstrate the effectiveness of the SHCS method.

In the actual work of reconnaissance sonar, without any prior information, an adaptive threshold selection method based on power spectrum entropy is proposed in this paper in order to solve this problem that is poor adaptability of pulse signal detection method. By using the difference between the power spectrum entropy of pulse signal and background noise, the background noise change is to find out. And the detection threshold value is dynamically adjusted, so as to realize that the detection threshold automatically changes with the background change. In this paper, the detection threshold, detection probability and false alarm probability of this method are deduced theoretically under the background of Gaussian white noise. The steps of automatic threshold selection under non-Gaussian background are presented. Through simulation analysis and sea trial data verification, the adaptability and practicability of this method are proved.

Adaptive array processing technology for a phased array radar is usually based on the assumption of a stationary environment; however, in real-world scenarios, nonstationary interference and noise deteriorate the performance of the traditional gradient descent algorithm, in which the learning rate of the tap weights is fixed, leading to errors in the beam pattern and a reduced output signal-to-noise ratio (SNR). In this paper, we use the incremental delta-bar-delta (IDBD) algorithm, which has been widely used for system identification problems in nonstationary environments, to control the time-varying learning rates of the tap weights. The designed iteration formula for the learning rate ensures that the tap weights adaptively track the Wiener solution. The results of numerical simulations show that in a nonstationary environment, the traditional gradient descent algorithm with a fixed learning rate has a distorted beam pattern and reduced output SNR; however, the IDBD-based beamforming algorithm, in which a secondary control mechanism is used to adaptively update the learning rates, showed a similar beam pattern and output SNR to a traditional beamformer in a Gaussian white noise background; that is, the main beam and null satisfied the pointing constraints, and the optimal output SNR was obtained. Although the proposed algorithm contains a matrix inversion operation, which has considerable computational complexity, this operation could be replaced by the Levinson–Durbin iteration due to the Toeplitz characteristic of the matrix; therefore, the computational complexity could be decreased to O(n), so additional computing resources are not required. Moreover, according to some intuitive interpretations, the reliability and stability of the algorithm are guaranteed.

A Bayesian estimation approach for enhancing speech signals which have been degraded by statistically independent additive noise is motivated and developed. In particular, minimum mean square error (MMSE) and maximum a posteriori (MAP) signal estimators are developed using hidden Markov models (HMMs) for the clean signal and the noise process. It is shown that the MMSE estimator comprises a weighted sum of conditional mean estimators for the composite states of the noisy signal, where the weights equal the posterior probabilities of the composite states given the noisy signal. The estimation of several spectral functionals of the clean signal such as the sample spectrum and the complex exponential of the phase is also considered. A gain-adapted MAP estimator is developed using the expectation-maximization algorithm. The theoretical performance of the MMSE estimator is discussed, and convergence of the MAP estimator is proved. Both the MMSE and MAP estimators are tested in enhancing speech signals degraded by white Gaussian noise at input signal-to-noise ratios of from 5 to 20 dB. >

In this study, formant tracking is investigated by refining the formants tracked by an existing data-driven tracker, DeepFormants, using the formants estimated in a model-driven manner by linear prediction (LP)-based methods. As LP-based formant estimation methods, conventional covariance analysis (LP-COV) and the recently proposed quasi-closed phase forward–backward (QCP-FB) analysis are used. In the proposed refinement approach, the contours of the three lowest formants are first predicted by the data-driven DeepFormants tracker, and the predicted formants are replaced frame-wise with local spectral peaks shown by the model-driven LP-based methods. The refinement procedure can be plugged into the DeepFormants tracker with no need for any new data learning. Two refined DeepFormants trackers were compared with the original DeepFormants and with five known traditional trackers using the popular vocal tract resonance (VTR) corpus. The results indicated that the data-driven DeepFormants trackers outperformed the conventional trackers and that the best performance was obtained by refining the formants predicted by DeepFormants using QCP-FB analysis. In addition, by tracking formants using VTR speech that was corrupted by additive noise, the study showed that the refined DeepFormants trackers were more resilient to noise than the reference trackers. In general, these results suggest that LP-based model-driven approaches, which have traditionally been used in formant estimation, can be combined with a modern data-driven tracker easily with no further training to improve the tracker’s performance.

Speech enhancement with Geometric Advent of Spectral subtraction using connected time-frequency regions noise estimation aims to de-noise or reduce background noise from the noisy speech for better quality, pleasantness and improved intelligibility. Numerous enhancement methods are proposed including spectral subtraction, subspace, statistical with different noise estimations. The traditional spectral subtraction techniques are reasonably simple to implement and suffer from musical noise. This study addresses the new approach for speech enhancement which has minimized the insufficiencies in traditional spectral subtraction algorithms using MCRA. This approach with noise estimation has been evolved with PESQ, the ITU-T standard; Frequency weighted segmental SNR and weighted spectral slope. The analysis shows that Geometric approach with time-frequency connected regions has improved results than old-fashioned spectral subtraction algorithms. The normal hearing tests has suggested that new approach has lower audible musical noise.

Over the past decades the problem of one channel, speech enhancement has been addressed by a great deal of researchers. In this work selected methods belonging to a variety of categories are applied to denoise speech signals corrupted by non-stationary urban noise. The performance of spectral subtraction, signal subspace, model-based and Kalman filtering approaches is evaluated. Several objective measures which are designed to predict human listening tests are employed in order to reach accurate conclusions. Two series of experiments were carried out while multiband spectral subtraction along with a short-time spectral amplitude (STSA) estimator based on the minimization of the mean square error (MSE) of the log-spectra are shown to outperform the rest of the algorithms.

Autoregressive HMM resolves biomolecular transitions from passive optical tweezer force measurements

Robust automatic speech recognition with missing and unreliable acoustic data

To allow users to access networks and services ubiquitously, more and more user equipments (UEs) are equipped with multiple radio transceivers. However, due to the proximity of transmit and receive frequencies of in-device wireless technologies, the transmitter (TX) of one in-device wireless technology can interfere with the reception of another wireless technology within the same device, and the spurious emissions of the aggressor TX may result in unacceptable interference at the victim receiver which is co-located in-device. Hence, it is crucial to detect the in-device interference and adapt the strategies to avoid harm. In this paper, we consider the scenario of LTE coexisting with WiFi. Different from traditional spectrum sensing in cognitive radio which detects signal in the background of Gaussian noise, the WiFi leakage detection problem in this paper aims to detect WiFi leakage in the background of LTE signal and Gaussian noise. This makes the problem challenging. By identifying unique features of the WiFi leakage, we investigate this problem thoroughly and propose two detection algorithms, including entropy ratio based and cyclostationary based algorithms. Simulation results under different scenarios are presented to illustrate the effectiveness and efficiency of the proposed detection techniques.

Keyword spotting on embedded platforms must balance accuracy and strict resource limits while remaining independent of network connectivity. The aim of the study was to develop and experimentally validate a classical, frugal recognition method that increases feature informativeness without increasing model complexity and is suitable for autonomous use on edge devices that rely only on a central processing unit. A weighted acoustic f ingerprinting mechanism was proposed. Mel-frequency cepstral coefficients, together with their derivatives, were reweighted, aggregated, and serialised into compact discrete “fingerprints”, which were then classified using the Levenshtein edit distance. Experiments were carried out on a Ukrainian-language command corpus from six native speakers (three male, three female), recorded with both headsets and far-field microphones; lexicons of 10, 100, and 200 words were evaluated under speaker-independent splits of 70%/15%/15%. The methodology comprised f ixed parametrisation of mel-frequency cepstral coefficients, construction of a static weighting vector, voice activity detection with spectral subtraction, uniform quantisation and serialisation, and deterministic edit-distance classification; for comparison, equal-weight baselines, hidden Markov models with Gaussian mixture emissions, Dynamic Time Warping, a lightweight convolutional neural network, and a reference depthwise-separable convolutional neural network were considered. The proposed method achieved macro-averaged harmonic means of precision and recall of 0.96/0.92/0.89 for 10/100/200-word lexicons in clean audio, and 0.78 at a signal-to-noise ratio of 5 decibels (100-word lexicon). The implementation required approximately 250 kilobytes of memory and operated with a real-time factor of 0.005 on Raspberry Pi 4 with 4 gigabytes, i.e., faster than real time. Superiority over equal-weight baselines, hidden Markov models with Gaussian mixture emissions, and Dynamic Time Warping was demonstrated, with performance approaching that of a compact convolutional neural network. It is concluded that weighted acoustic fingerprinting provides a robust, efficient, and autonomous keyword-spotting solution for deployments that use only a central processing unit