Postprocessing techniques for voice pitch trackers

Secrest, Bruce G.; Doddington, George R.

doi:10.1109/icassp.1982.1171719

Cited by 15 publications

(16 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The index j should be a value between 20 and 40 which could be obtained by divide the index i by k , where k belongs to the close zone [2,8]. The revise should be as follows:…”

Section: R I S T R I R I R I R I P I Elsementioning

confidence: 99%

“…As Figure1 to Figure4 show, in these algorithms, pitch is among the most important parameters in the final quality of the synthetic speech. Many pitch extraction algorithms have been proposed because of its importance [7][8][9][10][11][12]. However, these algorithms have some limitations because of some difficulties: the harmonic structure of speech signal varies widely; the head as well as end of a voiced period is difficult to judge; the pitch could change seriously some time itself.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Pitch Detection Algorithm for Ultra-low-bit-rate Vocoder

Li¹,

Jiang²,

Hao³

et al. 2012

Proceedings of the 2nd International Conference on Computer Application and System Modeling

View full text Add to dashboard Cite

Abstract-Pitch is one of the most important parameters to describe the speech characteristics. The paper proposed a new pitch detection algorithm for ultra-low-bit-rate vocoder based on dynamic programming as well as frequency-time domain analysis to eliminate the multi-half pitch errors. The algorithm detected the candidate pitches based on auto correlation function firstly and then utilized the dynamic programming and frequency-time domain analysis to get the final pitch. Simulation result showed that the algorithm had a better performance and could eliminate the multi-half pitch errors efficiently.

show abstract

“…The index j should be a value between 20 and 40 which could be obtained by divide the index i by k , where k belongs to the close zone [2,8]. The revise should be as follows:…”

Section: R I S T R I R I R I R I P I Elsementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A Pitch Detection Algorithm for Ultra-low-bit-rate Vocoder

Li¹,

Jiang²,

Hao³

et al. 2012

Proceedings of the 2nd International Conference on Computer Application and System Modeling

View full text Add to dashboard Cite

show abstract

“…Two approaches have been widely used. The first involves finding an optimal path through the F0 space over time by dynamically combining F0 salience values (also called measurement cost) and smoothness constraints (also called smoothness cost) using methods either based on the Viterbi algorithm [23] or dynamic programming (DP) [24], [25]. The second approach applies variants of the partial tracking (PT) algorithm, used in classical sinusoidal modeling [13], to forming multiple F0 trajectories/contours through the F0 candidate space over time.…”

Section: Melody Identificationmentioning

confidence: 99%

Vocal Melody Extraction in the Presence of Pitched Accompaniment in Polyphonic Music

Rao

2010

IEEE Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

Abstract-Melody extraction algorithms for single-channel polyphonic music typically rely on the salience of the lead melodic instrument, considered here to be the singing voice. However the simultaneous presence of one or more pitched instruments in the polyphony can cause such a predominant-F0 tracker to switch between tracking the pitch of the voice and that of an instrument of comparable strength, resulting in reduced voice-pitch detection accuracy. We propose a system that, in addition to biasing the salience measure in favor of singing voice characteristics, acknowledges that the voice may not dominate the polyphony at all instants and therefore tracks an additional pitch to better deal with the potential presence of locally dominant pitched accompaniment. A feature based on the temporal instability of voice harmonics is used to finally identify the voice pitch. The proposed system is evaluated on test data that is representative of polyphonic music with strong pitched accompaniment. Results show that the proposed system is indeed able to recover melodic information lost to its single-pitch tracking counterpart, and also outperforms another state-of-the-art melody extraction system designed for polyphonic music.Index Terms-Fundamental frequency estimation, music information retrieval (MIR), music transcription, predominant pitch detection.

show abstract

“…A global error correction routine is required for the pitch detection system to locate the incorrect estimates and correct the errors. Various techniques for global error correction have been proposed [4,11,14], ranging from Reddy's logical decision-based approach [11], to Secrest and Doddington's dynamic programming and pattern matching approach [14]. Linear and non-linear smoothing procedures are often used as global correction routines to enforce continuity of the pitch contour [4].…”

Section: Global Error Correctionmentioning

confidence: 99%

“…The F0 value from the reference laryngeal frequency contour is represented by Cepstrum pitch determination (CPD) [6] Feature-based pitch tracker (FBPT) [7] Harmonic product spectrum (HPS) [13] Integrated pitch tracking algorithm (IPTA) [14] Parallel processing method (PP) [3] Super resolution pitch determinator (SRPD) [5] Enhanced version of SRPD (eSRPD) [1] Modied AMDF-based PDA without error correction (mAMDF)…”

Section: Performance Comparison Of Pdasmentioning

confidence: 99%

A probabilistic approach to AMDF pitch detection

Ying

Jamieson

Michell

1996

Proceeding of Fourth International Conference on Spoken Language Processing ICSLP 96 ICSLP-96

View full text Add to dashboard Cite

We present a probabilistic error correction technique to be used with an average magnitude dierence function (AMDF) based pitch detector. This error correction routine provides a v ery simple method to correct errors in pitch period estimation. Used in conjunction with the computationally ecient AMDF, the result is a fast and accurate pitch detector. In performance tests on the CSTR (Center for Speech Technology Research) database, probabilistic error correction reduced the gross error rate from 6.07% to 3.29%.

show abstract

Postprocessing techniques for voice pitch trackers

Cited by 15 publications

References 1 publication

A Pitch Detection Algorithm for Ultra-low-bit-rate Vocoder

A Pitch Detection Algorithm for Ultra-low-bit-rate Vocoder

Vocal Melody Extraction in the Presence of Pitched Accompaniment in Polyphonic Music

A probabilistic approach to AMDF pitch detection

Contact Info

Product

Resources

About