Multi-pitch estimation

Christensen, Mads Græsbøll; Jakobsson, Andreas

doi:10.1016/j.sigpro.2007.10.014

Cited by 127 publications

(118 citation statements)

References 32 publications

Supporting

Mentioning

116

Contrasting

Unclassified

Order By: Relevance

“…Note that although any prior dependency between the complex amplitudes and the noise variance was included in the factorisation in (17), the dependency automatically appears through g. As reviewed in [47], the hyperparameter g can be set to a fixed value or treated as a random variable. When g is a random variable, the prior pdf of g can be derived from (26), (25), and (19) to…”

Section: The G-priormentioning

confidence: 99%

“…The more advanced algorithms are based on a signal model of the observed signal and are therefore referred to as parametric methods. These are typically maximum likelihood-based (ML) methods [16], [17], subspace-based methods [10], [18], filtering methods [19], [20], or Bayesian methods [7], [21], [22]. We refer the interested reader to [2] for a review of many of the non-Bayesian methods.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Default Bayesian Estimation of the Fundamental Frequency

Nielsen

Christensen

Jensen

2013

IEEE Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

Abstract-Joint fundamental frequency and model order estimation is an important problem in several applications. In this paper, a default estimation algorithm based on a minimum of prior information is presented. The algorithm is developed in a Bayesian framework, and it can be applied to both real-and complex-valued discrete-time signals which may have missing samples or may have been sampled at a non-uniform sampling frequency. The observation model and prior distributions corresponding to the prior information are derived in a consistent fashion using maximum entropy and invariance arguments. Moreover, several approximations of the posterior distributions on the fundamental frequency and the model order are derived, and one of the state-of-the-art joint fundamental frequency and model order estimators is demonstrated to be a special case of one of these approximations. The performance of the approximations are evaluated in a small-scale simulation study on both synthetic and real world signals. The simulations indicate that the proposed algorithm yields more accurate results than previous algorithms. The simulation code is available online.

show abstract

Section: The G-priormentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Default Bayesian Estimation of the Fundamental Frequency

Nielsen

Christensen

Jensen

2013

IEEE Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

show abstract

“…In this paper, we consider the pitch, , and the model order, , as known parameters. Numerous methods for estimation of these parameters exist [17], [18], [22]- [28]. Using Euler's formula, we can also write (2) as (3) where is the complex amplitude of the th harmonic, and denotes the elementwise complex conjugate of a matrix/vector.…”

Section: Signal Model and Problem Statementmentioning

confidence: 99%

“…Note that we only consider the effects of introducing non-causality in the filter designs and not of introducing non-causality in the estimation of the signal and noise statistics since the statistics are assumed to be known exactly in most parts of the paper. The proposed filter designs are based on two different decompositions of the desired signal; three designs are based on an orthogonal decomposition [12], and one is based on a harmonic decomposition [17], [18]. The orthogonal decomposition based filters are suitable for enhancing any kind of desired signal since they are designed using the noise statistics, whereas the harmonic decomposition based filter is calculated from the statistics of the desired signal under the assumption that it is periodic.…”

Section: Introductionmentioning

confidence: 99%

“…Moreover, we derive different non-causal orthogonal and harmonic decomposition based filters. Note that the causal filters proposed in [12], [17], [18], and [21] can be seen as special cases of the proposed designs. For the two particular cases where the filter is causal and anti-causal, respectively, we derive expressions for recursive updates of the orthogonal decomposition based filters and the maximum output signal-to-noise ratio of these.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Non-Causal Time-Domain Filters for Single-Channel Noise Reduction

Jensen

Benesty

Christensen

et al. 2012

IEEE Trans. Audio Speech Lang. Process.

Self Cite

View full text Add to dashboard Cite

Abstract-In many existing time-domain filtering methods for noise reduction in, e.g., speech processing, the filters are causal. Such causal filters can be implemented directly in practice. However, it is possible to improve the performance of such noise reduction filtering methods in terms of both noise suppression and signal distortion by allowing the filters to be non-causal. Non-causal timedomain filters require knowledge of the future, and are therefore not directly implementable. If the observed signal is processed in blocks, however, the non-causal filters are implementable. In this paper, we propose such non-causal time-domain filters for noise reduction in speech applications. We also propose some performance measures that enable us to evaluate the performance of non-causal filters. Moreover, it is shown how some of the filters can be updated recursively. Using the recursive expressions, it is also shown that the output SNRs of the filters always increase as we increase the length of the filter when the desired signal is stationary. From both the theoretical and practical evaluations of the filters, it is clearly shown that the performance of time-domain filtering methods for noise reduction can be improved by introducing non-causality.Index Terms-Linearly constrained minimum variance (LCMV) filter, maximum signal-to-noise ratio (SNR) filter, minimum variance distortionless response (MVDR) filter, noise reduction, non-causal filters, performance measures, time-domain filtering, Wiener filter.

show abstract

Musical Intelligence Analysis

Che

et al. 2012

Recent Advances in Computer Science and Information Engineering

View full text Add to dashboard Cite

Multi-pitch estimation

Cited by 127 publications

References 32 publications

Default Bayesian Estimation of the Fundamental Frequency

Default Bayesian Estimation of the Fundamental Frequency

Non-Causal Time-Domain Filters for Single-Channel Noise Reduction

Musical Intelligence Analysis

Contact Info

Product

Resources

About