Synthesis of the Singing Voice by Performance Sampling and Spectral Models

Bonada, Jordi; Serra, Xavier

doi:10.1109/msp.2007.323266

Cited by 65 publications

(51 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The most popular approach for singing synthesis is lyrics-tosinging (text-to-singing) synthesis where a user provides note-level score information of the melody with the corresponding lyrics to synthesize a singing voice [7][8][9]. To improve naturalness and provide original expressions, some systems [7] enable a user to adjust singing synthesis parameters such as pitch (F0) and dynamics (power).…”

Section: Vocalistener: Singing Synthesis System Imitating Pitch and Dmentioning

confidence: 99%

VocaListener and VocaWatcher: Imitating a human singer by using signal processing

Goto

Nakano

Kajita

et al. 2012

2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

In this paper, we describe three singing information processing systems, VocaListener, VocaListener2, and VocaWatcher, that imitate singing expressions of the voice and face of a human singer. VocaListener can synthesize natural singing voices by analyzing and imitating the pitch and dynamics of the human singing. VocaListener2 imitates temporal timbre changes in addition to the pitch and dynamics. In synchronization with the synthesized singing voices, VocaWatcher can generate realistic facial motions of a humanoid robot, the HRP-4C, by analyzing and imitating facial motions of a human singing that are recorded by a single video camera. These systems that focus on "imitation" are not only promising for representing human-like naturalness, but also useful for providing intuitive control means.

show abstract

Section: Vocalistener: Singing Synthesis System Imitating Pitch and Dmentioning

confidence: 99%

VocaListener and VocaWatcher: Imitating a human singer by using signal processing

Goto

Nakano

Kajita

et al. 2012

2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…Altogether, the error function (10) can be viewed as a function of the union of all Y (k) and g(t). If the g(t) are fixed, then the error function has a unique minimum due to (9). On the other hand, if the Y (k) are fixed, then it has also a unique minimum because (11) has only one solution.…”

Section: Optimal Window Functionmentioning

confidence: 99%

“…However, it is possible to develop an iterative method. By alternatingly calculating the optimal Y from g by (9) and g from Y by (11), we get a scheme…”

Section: Optimal Window Functionmentioning

confidence: 99%

See 1 more Smart Citation

Optimized Sinusoid Synthesis via Inverse Truncated Fourier Transform

Kutil

2009

IEEE Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

Abstract-It was shown that sinusoid synthesis can be implemented efficiently by an inverse Fourier transform on consecutive frames where all but a small number of coefficients per oscillator are dropped. This leads to a compromise between computational complexity and approximation accuracy. The method can be improved by two approaches. First, optimal coefficients can be found by minimizing the average approximation error. Second, the optimal window function can be found through an iterative process. The gain in signal-to-noise ratio (SNR) is between 10 and 40 dB and can be used to reduce computational complexity while satisfying required synthesis quality.

show abstract

“…Most related work [4,5] investigate the differences between singing and speech. Some studies develop methods for singing voice synthesis [6,7], and some discuss how to convert speech into singing [8], according to the specified melody. This work investigates two SNID systems trained using spoken data.…”

Section: Introductionmentioning

confidence: 99%

Automatic Singer Identification Based on Speech-Derived Models

Tsai¹,

Lee²

2012

IJFCC

View full text Add to dashboard Cite

Abstract-Automatic singer identification (SNID) aims to determine who among a set of singers perform a given music recording. So far, most existing SNID methods follow a framework stemming from speaker identification (SPID) research, which models each person's characteristics using his/her voice data. This framework, however, is impractical in many SNID applications, because acquiring solo a cappella from each singer is usually not as feasible as collecting spoken data from each speaker in SPID applications. In view of the easy availability of spoken data, this work investigates the possibility of modeling singers' voices using spoken data instead of singing data. However, our experiment found it difficult to replace singing data fully by using spoken data in singer voice modeling, due to the significant difference between singing and speaking for most people. Thus, we propose an alternative solution based on the use of few available singing data. The idea is to modify speech-derived voice models using MAP adaptation on few singing data, so that the adapted voice models can cover performers' singing characteristics. Our experiments found that most of the singing clips can be correctly identified using the adapted voice models.Index Terms-Model adaptation, singer identification, speaker identification.

show abstract

Synthesis of the Singing Voice by Performance Sampling and Spectral Models

Cited by 65 publications

References 24 publications

VocaListener and VocaWatcher: Imitating a human singer by using signal processing

VocaListener and VocaWatcher: Imitating a human singer by using signal processing

Optimized Sinusoid Synthesis via Inverse Truncated Fourier Transform

Automatic Singer Identification Based on Speech-Derived Models

Contact Info

Product

Resources

About