Literatura científica selecionada sobre o tema "Singing voice transformation"

This work reports development of a MIDI-to-Singing song synthesis that will produce audio files from MIDI data and arbitrary Romaji lyrics in Japanese. The MIDI-to-Singing system relies on the Flinger (Festival singer) for singing voice synthesis. Originally, this MIDI-to-Singing system was developed by English. Based on some Japanese pronunciation rules, a Japanese MIDI-to-Sing synthesis system was developed and derived. For a language transfer like Festival synthesized singing, two major tasks are the modifications of a phoneset and a lexicon. Originally, MIDI-to-Sing song synthesis can create singing voices in many languages, but there is no existing Japanese festival diphone voice available right now. We therefore used a voice transformation model in festival to develop Japanese MIDI-to-Singing synthesis. An evaluation of a song listening experiment was conducted and the result of this voice conversion showed that the synthesized singing voice successfully migrate from English to Japanese with high voice quality.

Adolescence is characterised by a distinctive phase of vocal development which bridges childhood and adulthood. Various research studies over the past twenty years have demonstrated that the physical maturation of the adolescent voice mechanism produces a systematic change in both the male and female singing voice. Longitudinal research data indicate that there are distinctive features to each stage of adolescent voice change and that, with an appropriate matching of repertoire, it is possible for all young people to continue to sing successfully throughout this period. The traditional and stereotypical notion that adolescent male voices ‘break’ is untenable in the light of this research evidence and it is suggested that a concept of adolescent singing voice ‘transformation’ or ‘change’ is a more accurate representation of the physiological reality. It is a weakness of the revised National Curriculum for Music (1995) that it makes no appropriate reference to this unique period of adolescent voice change and, as a result, teachers receive inadequate statutory guidance on the development of singing at Key Stage 3.

In this publication, we present a deep learning-based method to transform the f0 in speech and singing voice recordings. f0 transformation is performed by training an auto-encoder on the voice signal’s mel-spectrogram and conditioning the auto-encoder on the f0. Inspired by AutoVC/F0, we apply an information bottleneck to it to disentangle the f0 from its latent code. The resulting model successfully applies the desired f0 to the input mel-spectrograms and adapts the speaker identity when necessary, e.g., if the requested f0 falls out of the range of the source speaker/singer. Using the mean f0 error in the transformed mel-spectrograms, we define a disentanglement measure and perform a study over the required bottleneck size. The study reveals that to remove the f0 from the auto-encoder’s latent code, the bottleneck size should be smaller than four for singing and smaller than nine for speech. Through a perceptive test, we compare the audio quality of the proposed auto-encoder to f0 transformations obtained with a classical vocoder. The perceptive test confirms that the audio quality is better for the auto-encoder than for the classical vocoder. Finally, a visual analysis of the latent code for the two-dimensional case is carried out. We observe that the auto-encoder encodes phonemes as repeated discontinuous temporal gestures within the latent code.

The relevance of the topic. The dynamics of cultural development in the 21st century sets for the interpreter-vocalist tasks that require a quick and high-quality response to public inquiries (challenges of the time), on which the performing demand at a particular institution directly depends. MC “The Kharkiv Regional Philharmonic Society” is a concert organization, where, unlike the opera theatre, there are not clearly regulated genre priorities. Therefore, opera, jazz and even pop music organically coexist on the same stage. This genre and style diversity entails the need for the soloist to comply with the performing complex, which includes a number of requirements: the vocal-technical and stage freedom, the mastery of the art of improvisation (working with the audience), the vocal endurance, taking into account the repertory load (during one hour and a half). Last but not least, there are such criteria as the vocal-performing aesthetics, the ability for gender-performing transformation and the vocal-figurative reincarnation in one concert, etc. The purpose of the article is to substantiate the role of the phonopedic method for the development of a singing voice for developing the vocal-performing universalism in the conditions of the competition between genre-performing styles and the trends of the modern vocal culture. Analysis of recent research and publications. A great deal of interesting research and biographical works (I. Arkhipova, P. Domingo, S. Lemeshev, E. Nesterenko, B. Hmyrya, T. Madysheva,) is devoted to studying the specifics of the vocal-performing interpretation. The practicing teachers and scientists (L. Dmitriev, J. Lauri-Volpi, I. Nazarenko, V. Yushmanov) reveal the performing secrets, share their stage experience, meticulously identifying the nature and the possibilities of the voice apparatus, make historical excursions and acquaint with the vocal traditions of the past. However, none of the above scientific sources presents a systematic picture reflecting the vocal-performing process as a complex phenomenon. The author sees the indicated problem as relevant and requiring a multilateral discussion of specialists. The presentation of the main material. As of today, the professional activity of a vocal artist, regardless of the genre direction, faces very high requirements, the most important of which is the interpretative universalism. Its fundamental element is the vocal technique, the possession of which is the basic condition that ensures the singer a long and healthy professional life. However, up to the present, it is the vocal technique that continues to be an area of the open problem of the singing instrument, as a sound-forming object used by vocal performers to realize their artistic intentions and tasks. The key issue of the vocal method and one of the fundamental questions of the theory of the singing art, the psychophysics that controls the phonation process remains its unexplored area. According to V.I. Yushmanov, the technically perfect singing (or the vocal school) is a factor by which, first of all, the vocal mastery is evaluated, and the ability to sing, maintaining the phonetic clarity of vocal speech in combination with timbre rich, and at the same time bright, flying sound of the voice, steady dynamically and in the pitch on a range of at least two octaves – is one of the main requirements of the singing profession, the very necessary condition that provides the singer with the opportunity to realize their artistic intentions. But at the same time, the impossibility of separating the singing instrument from the singer himself/herself and the secrecy of the work of the control system of the singing process initially caused the emergence of the persistent illusion that, unlike instrumentalists, singers deal with the voice – perceived only acoustically (by the hearing). In this regard, the development of the vocal technique, called the “singing voice setting” must include a detailed study of the functional structure of the vocal apparatus, as a musical instrument, together with the psychophysical peculiarities. It should be noted that in the gender (male and female) and age (children, adolescents, adults) senses, the psychophysical peculiarities of the management of the vocal phonation process are diametrically opposed and require an individual approach in the vocal pedagogy. The methods of influence on psychophysics, which are responsible for managing the vocal phonation process, the principles of its flexible correction are presented in detail in the field of phonopedia and are successfully applied by vocal phonopedists. Unfortunately, the majority of the vocal-performing and phonopedic practices coexist in parallel and intersect only in critical cases (in case of functional disorders of the vocal apparatus). For ten years of my performing on the stage of the concert organization MC “The Kharkiv Regional Philharmonic Society” I have had a unique opportunity to try myself in different vocal genres and vocally reincarnate within the fames of even one concert. Such working conditions discipline, force to develop, and provide the opportunity to experiment and learn the professional foundations of singing with an even new power. And, of course, it would be impossible without the proper vocal and technical foundations, which should be constantly replenished and improved thanks to the knowledge of the phonopedic method of the voice development. The conclusion. On the example of the personal experience in chamber and concert performances it has been shown that the presented issues are relevant and deserve attention. The “tandem” of phonopedia and vocal theory not only opens new horizons for singers in their vocal-performing practice, but also shows the shortest way to mastering the singing and interpretive universalism.

The article presents the results of the research on musical and artistic creation of the image of one of the most striking creatures — the cuckoo. The ways to implement the peculiarities of its singing in instrumental compositions of the XVIIth – XIXth centuries are revealed and described in cultural and stylistic context. The authors analyze the role of timbre-intonational components of the bird's singing in the formation of some acoustic model, which becomes the basis for its representation by instrumental means in composer's works. The special sound pitch and rhythmic organization, together with timbre, representing the characteristics of the acoustic signal source, make it possible to recognise the cuckoo's voice in nature, while measured and repeated intonational complex becomes an absolute index-sign for perception. This research revealed that the method of artistic modeling appears to be the main one in creating the image of a cuckoo in music. A two-step falling intonational figure, which migrates in the works of composers of different styles and nationalities, becomes the invariant. Preserving its main features, it changes under the influence of the context — metrorhythmic, dynamic, agogic and other components of the text. Distinguishing the characteristic intonation in our perception, the timbre component largely determines its transformation in a musical text. The result is a wide development of timbre-intonational model.

The gramophone record industry had developed in the Yugoslav region from the beginning of the 20th century. The paper is based on an analysis of the corpus of 78 rpm records of singing with gusle accompaniment, which were produced between 1908 and 1932. Available recordings highlight the issue of representing both the epic and the gusle playing tradition in this media format and its relationship with ?unmediated? live gusle playing practice. Therefore the authors opted to analyze gramophone records as both a text in culture and an actor in tradition. After introductory theoretical and methodological remarks, the authors offer a brief description of the historical, political and socio-cultural context that emphasized epic singing with gusle accompaniment as a representative traditional genre in this area. For that reason it was noteworthy for both western and local production companies which made recordings in the Balkans (Deutsche Grammophon Gesellschaft mbH., Odeon Records, Marsh Laboratories, Inc., Edison Bell Penkala Ltd.). Analysis of the recordings is focused on examination of the way in which gusle players responded to different requirements of the new media (e.g. the insufficient capacity of a record itself compared to the usual duration of a performance; the reduction of a complex form of artistic communication to an oral, auditory message, while additional forms of non-verbal communication are excluded). Through discussion of the treatment of verbal and musical components of the recorded performances it has been shown that tradition was simultaneously exemplified and reshaped by this new medium. In addition to the guslars themselves, being already recognized artists in this traditional genre and the acoustic source (the voice accompanied by the gusle), the representative base of the epic tradition comprised traditional (poetic) texts, although modifications / innovations are recognizable at different levels of verbal content, as well as on the level of music interpretation. On these bases, it is possible to talk about the contribution of new media to the professionalization of guslars? practice and the creation of ?stars? among them on the one hand, and the progressive transformation of once-active audience members (in the sense of potentially exchangeable performer / listener positions) into passive buyers and consumers, on the other. It is noted that these, first gramophone records of guslars had a great role in and impact on the survival of the epic singing tradition, brokering its promotion in urban areas and among the ?cultural elite?. Finally, in this way they contributed to the strengthening of this tradition in a historical period that brought disintegration of the system of traditional culture and ?crisis? of the most of the current classic folklore genres.

Background. The modern science reconsiders various conceptions, which were influencing the theory and practice of musical art over the centuries. Particularly, there is much talk today about the fact that marking of female opera roles as “coloratura” according to the principle of their technical complexity and diapason wideness is quite nominal and not connected directly with singing voices’ gradation. Gradually entrenched tendency of denial of the female voice’s definition as “coloratura” has developed, and it is based on the argument that this characteristic reflects parameters of composer’s objectives rather than the voice’s nature. Probably, that’s why there are works in repertoire of certain female vocalists (for instance, Maria Callas and contemporary Canadian singer Natalie Choquette), which are usually performed by owners of “different” voices. However one cannot deny the fact that certain opera roles are composed specifically for coloratura soprano, despite the fact that indications of it are missing in manuscripts. N. Rimsky-Korsakov’s opera heritage is exponential in this connection; the destination of particular female roles for coloratura soprano is unquestionable – Snow Maiden, Marfa, Volkhova, Tsaritsa of Shemakha, etc. And though this roles are performed by female vocalists of various voices in today’s theatrical practice, it seems to us that voice’s characteristics have principal significant for appropriate implementation of author’s conception. Objectives. Thus, the purpose of the study is to identify the significance of the voice’s particularity factor as a carrier of a certain imagery in the composer’s conception and in the performer embodiments of opera parts (separated opera arias). Methods. The methodological basis of the study is the unity of scientific approaches, among which the most important is a functional one, associated with the analysis of the genre as a typical structure. Results. The Gogol’s plot became the basis of several operas, the most famous of which are “Cherevichki” (“The Little Shoes”) by P. Tchaikovsky and “Noch’ pered Rozhdestvom” (“Christmas Eve”) by N. Rimsky Korsakov. The comparison of this two works clearly shows the fundamental difference in composers’ conceptions. In P. Tchaikovsky’s interpretation the lyrical line is extracted from the literary primary source. But for N. Rimsky Korsakov the comparison of the real and fantastic world becomes the main thing in this opera. Therefore the role of Vakula is written quite schematically, but Oksana’s image is interesting developed, it is presented in progress – from carefree girl to loving woman. This progress is obvious when comparing Oksana’s Arias from Act I and Act IV. The Aria from Act I is a peculiar synthesis of national and Italian singing traditions, a prime example of entrance aria (di sortita), which presents the character’s portrayal and comprises such basic components as slow introduction and episodes demonstrating technical possibilities of the voice. Oriental intonations, which are specific for composer’s vocal works, coupled with coloraturas, give the impression that Oksana is “not from this world”. According to lots of researchers, the whole N. Rimsky Korsakov’s opera “metacycle” is an artistic integrity united by a generic idea that defined the unity of approach to implementing of “type” (including female) characters. The intonational canvas of every particular role (the choice of so-called intonational complexes – “a cold” or “a warm”) is determined by character’s affiliation with natural or fairy-tale locus. Oksana’s portrayal for N. Rimsky Korsakov has been ambivalent. On the one hand, she doesn’t relate to “another world” like Snow Maiden or Volkhova; on the other hand – Oksana is a fairy-tale character. Therefore composer uses partly the same strokes in the development of the portrayal as for female fairy-tale characters. However, the formation of this character’s personality is revealed through the transformation of “cold” (fairy-tail) intonational complex to “warm” (“alive”). Two performances of the first Oksana’s Aria are briefly reviewed as an example: the concert performance by Gohar Gasparyan, an Armenian lyrical coloratura soprano (1924–2007), and a recording from the Inessa Prosalovskaya’s CD “Arias from operas”, the Russian lyrical dramatic soprano (born in 1947). G. Gasparyan’s idea was to present the portrayal of a young girl of the people. Therefore the singer levels virtuosic components of music material as much as possible, and a coloring of her voice, for which it was easy to sing the second A sharp above middle C, emphasizes lyrical hints of Oksana’s Aria. Also the significant textual cuts becomes one of important parameters of creation of a “gentle” young girl’s portrayal; they not only transform the expanded aria into the form, which is close to a song in scale, but also significantly reduce specifically those snippets, in which technical difficulties are concentrated. Version by I. Prosalovskaya presents another interpretation, original sound of which is largely due to the singer’s timbre of voice. Its deepness, expressivity and completion absolutely modify personal characteristics of the N. Rimsky-Korsakov’s character. Therefore we observe not a young girl already, but a woman – passionate and confident. Thus, it could be concluded that timbre color’s specificity of the voice of female opera singer has a significant impact on features of the character that she embodies. It is obvious that this specificity determines all the parameters of the performer’s version of the composer’s work (both a separate aria and the opera as a whole). A more detailed study of the relationship between the voice timbre and the semantic and compositional decisions characterizing an individual performer style seems to us a promising direction for further research.

Background. The contemporary choral art has accumulated many bright samples, which present various performing directions and genres. Particularly interesting are those that include the traits traditionally not inherent to the choral performance. We will consider some of the requirements for the performance of choral music, put forward by the composers of the twentieth century, in particular, representatives of the new Viennese school and the Italian avant-garde, as well as the performance conditions, which require the tendency to adaptation for the stage of choral works and the improvisational component of musical thinking today. The need for such analysis is connected with vocal-performing training of contemporary choirmasters. The choral collective is a complex musical organism, and its successful work depends not only on the conductor, but also on the performers, therefore, there is a question of improving the complex development of the vocal-technical skills of the choirmaster, taking into account the specifics of his/her future activities. Methods. The scientific and methodological basis of improving the performing skills must be deepened in the issues of the psychology of performance and expanded with the questions of combining the choral performance with theatre. Due to such a methodological basis, it is possible to study extremely complex psychological processes of artistically meaningful vocal intonation, formation, control and implementation of performing skills, peculiarities of the psychological state of the choir performers, etc. The article is based on the analysis of materials on the development of the choral performance in various artistic directions and their synthesis (Kyrylenko, Ya., 2017; Ryzhynsky, A. 2015, 2016; Saponov, M., 1982), as well as the study of psychophysiological regularities of the vocal education in different performance conditions (Lukishko, A. 1984; Morozov, V., 2008, Selezneva, N., 2005). Objectives. The purpose of the article is to consider the performance tasks in the contemporary choral art and to outline the factors that shape the skills for the implementation of these tasks. Results. The analysis of the development of the vocal-choral art proves that vocal-technical means are defined and formed in unity with the performing tasks that characterize certain musical trends, styles or genres. Throughout the 20th century, in the choral art appears the works that became the result of the composers’ search for new versions of the timbre sound. The representatives of the new Viennese school A. Schoenberg and A. Webern were the first to overcome the timbre “conservatism” of the choir. In the choral creative activities by A. Schoenberg, the experiments with the use of timbre can be found. Namely, the use of expressiveness of the unison of the alt and tenor and the falsetto singing of the tenor, which provided for the alignment of the timbre and corrected dynamic misbalance of the artificial ensemble of tenor and soprano. However, the further introduction of falsetto singing was no longer a technical necessity, but became a means of a new expressiveness. The experiments are typical for the choral creativity by A. Webern, where the following two tendencies are clearly distinguished: 1) the jump-like structure of the melody in combination with syllabics; 2) the presence of pauses, which divide the motive into separate intonations. The purpose of the texture transformation was to uncover a fundamentally new sound of the choir, the creation of a new choral timbre. A. Webern achieves the timbre richness through a constant game with various nuances, strokes, and juxtaposition of the singing registers. The Italian composers L. Nono (1924–1990), B. Maderna (1920–1973) and L. Berio (1925–2003) became the followers of Webern’s creative experiments with regard to the organization of sound, the texture of the composition and timbres. In their creative work, they showed the possibility of an organic combination of centuries-old traditions of the vocal art with the latest composition techniques. In the modern choral art, a great interest is paid to theatrical actions during the choral performance. Hence, the new performing tasks rise for both, the conductor and performers of the choir. So, in the dramatized choral performance, a personality of an actor-singer is characterized by a certain complex of skills. The latter include ownership of different types of expressive intonation: vocal intonation, the verbal intonation and the acting intonation or the intonation of the gesture. The choral performer’s instrument is a singing voice. The vocal development of choral singers bases on the objective laws of the singing, the basis for understanding of which is the positions of musical acoustics. The choirmasters need to know, which acoustic patterns influence the formation of the singing voice in the choir, which of its properties develop automatically, and which require special techniques that stimulate an individual development. Another important aspect on which it is necessary to focus attention in the modern vocal education is the improvisational thinking, the ability to creative ingenuity. The tasks of the modern choral art are the preservation, the development, and sometimes the revival of the traditions of improvisational vocal-ensemble music. The solutions of these tasks one should look for in several directions. The first of them is the preservation of the traditions of Western European professional polyphonic (counterpoint) music, which is determined by the highest level of combination of the canonically invariable melody, of the cult or secular, with improvisational melodic decorations. The second is due to the preserving the traditions of collective vocal improvisation in folk singing cultures. The third direction is due to performance tasks, which are provided by the composers’ creativity of contemporary authors. Conclusion. The contemporary choirmaster needs to have a wide range of knowledge in the questions of history and theory of the world music, choral science, vocal pedagogy, acting skills, psychology, without which it is impossible to solve numerous issues that constantly arise in practice. The choirmaster must develop a critical and analytical thinking in him/her, be able to judge right the work of his/her mind and be thoughtful in the actions. The creative approach in work contributes to the flexibility in solving any problems – perhaps, it allows to abandon the usual techniques in work, which at the present moment do not produce the desired results. Practice is the only criterion for the truth of the selected actions.

Among the early medieval authors, Amalarius of Metz (8th–9th century) is one of those who discussed the impact of religious music and song on the body and soul. In his main work: the Liber Officialis, listening and singing liturgical music are depicted as having a corporeal effect that generates sensations of an intense sensory and emotional character. In Amalarius, living the musical religious phenomenon not only coincides with the idea that music can evoke emotions but there is something that goes further. What Amalarius emphasizes is a particular emotion: a “spiritual state” of the nakedness of the heart, almost a weakness of those who are capable of tears and sensitive to God’s voice. During the patristic era, especially in the East, “penthos” (compunction) was used to describe the experience of tears in prayer or meditation; however, Fathers of the Church described liturgical music as an obstacle to compunction. For this reason, an evolution of that compunctory doctrine emerges from the exposition of Amalarius. In this context, it is not a question of crying for one’s sins but of exploiting a natural power (vis) of music. By simply listening to music, a person would seem capable of being moved and reaching a particularly “receptive state” to welcome the Word of God and make it bear fruit. What Amalarius describes in religious music seems to be the natural experience that one feels when, just listening to a melody, a tear spontaneously falls. This physical reaction is connected to a spiritual transformation that seems to pass through the flesh (carnalia) of our humanity. The result of this singular experience, strongly connected to musical ethics, is the conversion to good action and the possibility to dispose human beings to attentive, deep, and fruitful listening.

Le but de cette thèse était de conduire des recherches sur la synthèse et transformation expressive de voix chantée, en vue de pouvoir développer un synthétiseur de haute qualité capable de générer automatiquement un chant naturel et expressif à partir d’une partition et d’un texte donnés. 3 directions de recherches principales peuvent être identifiées: les méthodes de modélisation du signal afin de générer automatiquement une voix intelligible et naturelle à partir d’un texte donné; le contrôle de la synthèse, afin de produire une interprétation d’une partition donnée tout en transmettant une certaine expressivité liée à un style de chant spécifique; la transformation du signal vocal afin de le rendre plus naturel et plus expressif, en faisant varier le timbre en adéquation avec la hauteur, l’intensité et la qualité vocale. Cette thèse apporte diverses contributions dans chacune de ces 3 directions. Tout d’abord, un système de synthèse complet a été développé, basé sur la concaténation de diphones. L’architecture modulaire de ce système permet d’intégrer et de comparer différent modèles de signaux. Ensuite, la question du contrôle est abordée, comprenant la génération automatique de la f0, de l’intensité, et des durées des phonèmes. La modélisation de styles de chant spécifiques a également été abordée par l’apprentissage des variations expressives des paramètres de contrôle modélisés à partir d’enregistrements commerciaux de chanteurs célèbres. Enfin, des investigations sur des transformations expressives du timbre liées à l'intensité et à la raucité vocale ont été menées, en vue d'une intégration future dans notre synthétiseur
This thesis aimed at conducting research on the synthesis and expressive transformations of the singing voice, towards the development of a high-quality synthesizer that can generate a natural and expressive singing voice automatically from a given score and lyrics. Mainly 3 research directions can be identified: the methods for modelling the voice signal to automatically generate an intelligible and natural-sounding voice according to the given lyrics; the control of the synthesis to render an adequate interpretation of a given score while conveying some expressivity related to a specific singing style; the transformation of the voice signal to improve its naturalness and add expressivity by varying the timbre adequately according to the pitch, intensity and voice quality. This thesis provides some contributions in each of those 3 directions. First, a fully-functional synthesis system has been developed, based on diphones concatenations. The modular architecture of this system allows to integrate and compare different signal modeling approaches. Then, the question of the control is addressed, encompassing the automatic generation of the f0, intensity, and phonemes durations. The modeling of specific singing styles has also been addressed by learning the expressive variations of the modeled control parameters on commercial recordings of famous French singers. Finally, some investigations on expressive timbre transformations have been conducted, for a future integration into our synthesizer. This mainly concerns methods related to intensity transformation, considering the effects of both the glottal source and vocal tract, and the modeling of vocal roughness

Le but de cette thèse était de conduire des recherches sur la synthèse et transformation expressive de voix chantée, en vue de pouvoir développer un synthétiseur de haute qualité capable de générer automatiquement un chant naturel et expressif à partir d’une partition et d’un texte donnés. 3 directions de recherches principales peuvent être identifiées: les méthodes de modélisation du signal afin de générer automatiquement une voix intelligible et naturelle à partir d’un texte donné; le contrôle de la synthèse, afin de produire une interprétation d’une partition donnée tout en transmettant une certaine expressivité liée à un style de chant spécifique; la transformation du signal vocal afin de le rendre plus naturel et plus expressif, en faisant varier le timbre en adéquation avec la hauteur, l’intensité et la qualité vocale. Cette thèse apporte diverses contributions dans chacune de ces 3 directions. Tout d’abord, un système de synthèse complet a été développé, basé sur la concaténation de diphones. L’architecture modulaire de ce système permet d’intégrer et de comparer différent modèles de signaux. Ensuite, la question du contrôle est abordée, comprenant la génération automatique de la f0, de l’intensité, et des durées des phonèmes. La modélisation de styles de chant spécifiques a également été abordée par l’apprentissage des variations expressives des paramètres de contrôle modélisés à partir d’enregistrements commerciaux de chanteurs célèbres. Enfin, des investigations sur des transformations expressives du timbre liées à l'intensité et à la raucité vocale ont été menées, en vue d'une intégration future dans notre synthétiseur
This thesis aimed at conducting research on the synthesis and expressive transformations of the singing voice, towards the development of a high-quality synthesizer that can generate a natural and expressive singing voice automatically from a given score and lyrics. Mainly 3 research directions can be identified: the methods for modelling the voice signal to automatically generate an intelligible and natural-sounding voice according to the given lyrics; the control of the synthesis to render an adequate interpretation of a given score while conveying some expressivity related to a specific singing style; the transformation of the voice signal to improve its naturalness and add expressivity by varying the timbre adequately according to the pitch, intensity and voice quality. This thesis provides some contributions in each of those 3 directions. First, a fully-functional synthesis system has been developed, based on diphones concatenations. The modular architecture of this system allows to integrate and compare different signal modeling approaches. Then, the question of the control is addressed, encompassing the automatic generation of the f0, intensity, and phonemes durations. The modeling of specific singing styles has also been addressed by learning the expressive variations of the modeled control parameters on commercial recordings of famous French singers. Finally, some investigations on expressive timbre transformations have been conducted, for a future integration into our synthesizer. This mainly concerns methods related to intensity transformation, considering the effects of both the glottal source and vocal tract, and the modeling of vocal roughness

Aquesta tesi doctoral versa sobre el processament digital de la veu cantada, més concretament, sobre l'anàlisi, transformació i síntesi d'aquets tipus de veu en el domini espectral, amb especial èmfasi en aquelles tècniques rellevants per al desenvolupament d'aplicacions musicals.

La tesi presenta nous procediments i formulacions per a la descripció i transformació d'aquells atributs específicament vocals de la veu cantada. La tesis inclou, entre d'altres, algorismes per l'anàlisi i la generació de desordres vocals como ara rugositat, ronquera, o veu aspirada, detecció i modificació de la freqüència fonamental de la veu, detecció de nasalitat, conversió de veu cantada a melodia, detecció de cops de veu, mutació de veu cantada, i transformació de veu a instrument; exemplificant alguns d'aquests algorismes en aplicacions concretes.
Esta tesis doctoral versa sobre el procesado digital de la voz cantada, más concretamente, sobre el análisis, transformación y síntesis de este tipo de voz basándose e dominio espectral, con especial énfasis en aquellas técnicas relevantes para el desarrollo de aplicaciones musicales.

La tesis presenta nuevos procedimientos y formulaciones para la descripción y transformación de aquellos atributos específicamente vocales de la voz cantada. La tesis incluye, entre otros, algoritmos para el análisis y la generación de desórdenes vocales como rugosidad, ronquera, o voz aspirada, detección y modificación de la frecuencia fundamental de la voz, detección de nasalidad, conversión de voz cantada a melodía, detección de los golpes de voz, mutación de voz cantada, y transformación de voz a instrumento; ejemplificando algunos de éstos en aplicaciones concretas.
This dissertation is centered on the digital processing of the singing voice, more concretely on the analysis, transformation and synthesis of this type of voice in the spectral domain, with special emphasis on those techniques relevant for music applications.

The thesis presents new formulations and procedures for both describing and transforming those attributes of the singing voice that can be regarded as voice specific. The thesis includes, among others, algorithms for rough and growl analysis and transformation, breathiness estimation and emulation, pitch detection and modification, nasality identification, voice to melody conversion, voice beat onset detection, singing voice morphing, and voice to instrument transformation; being some of them exemplified with concrete applications.

La voix humaine est une grande source de fascination et un objet de recherche depuis plus de 100 ans. Pendant ce temps, de nombreuses technologies ont germées autour de la voix, comme le vocodeur, qui fournit une représentation paramétrique de la voix, couramment utilisée pour la transformation de la voix. Dans cette tradition, les limites des approches basées uniquement sur le traitement du signal sont évidentes : Pour créer des transformations cohérentes, les dépendances entre les différentes propriétés vocales doivent être bien comprises et modélisées avec précision. Modéliser ces corrélations avec des heuristiques obtenues par des études empiriques ne suffit pas à créer des résultats naturels. Il est nécessaire d'extraire systématiquement des informations sur la voix et d'utiliser automatiquement ces informations lors du processus de transformation. Les progrès récents de la puissance de calcul permettent cette analyse systématique des données au moyen de l'apprentissage automatique. Cette thèse utilise donc l'apprentissage automatique pour créer un système neuronal de transformation de la voix. Le système neuronal de transformation de la voix, présenté ici, fonctionne en deux étapes : Tout d'abord, un vocodeur neuronal permet d'établir une correspondance entre la forme d'onde et une représentation mel-spectrogramme des signaux vocaux. Ensuite, un auto-encodeur avec un goulot d'étranglement permet de démêler différentes propriétés de la voix du reste de l'information. L'auto-encodeur permet de modifier une propriété de la voix tout en ajustant automatiquement d'autres caractéristiques de façon à en conserver le réalisme. Dans la première partie de cette thèse, nous comparons différentes approches du vocodage neuronal et nous expliquons pourquoi la représentation mel-spectrogramme est plus adapté pour la transformation neuronale de la voix plutôt que les espaces paramétriques du vocodeur conventionnels. Dans la deuxième partie, nous présentons l'auto-encodeur avec goulot d'étranglement de l'information. L'auto-encodeur crée un code latent indépendant du conditionnement en entrée. En utilisant ce code latent, le synthétiseur peut effectuer la transformation en combinant le code latent original avec une courbe de paramètres modifiée. Nous transformons la voix en utilisant deux paramètres de contrôle : la fréquence fondamentale et le niveau sonore vocal. La transformation de la fréquence fondamentale est un problème qui a longtemps été abordé : Notre approche est comparable aux techniques existantes puisqu'elles utilisent la fréquence fondamentale comme paramètre. Cela nous permet également d'étudier comment l'auto-encodeur modélise les dépendances entre la fréquence fondamentale et d'autres propriétés de la voix dans un environnement connu. Quant au niveau sonore vocal, nous sommes confrontés au problème de la rareté des annotations. Par conséquent, nous proposons d'abord une nouvelle technique d'estimation du niveau sonore vocal dans de grandes bases de données de voix ; puis nous utilisons ces annotations pour entraîner un auto-encodeur avec goulot d'étranglement permettant de modifier le niveau sonore vocal
Human voice has been a great source of fascination and an object of research for over 100 years. During that time numerous technologies have sprouted around the voice, such as the vocoder, which provides a parametric representation of the voice, commonly used for voice transformation. From this tradition, the limitations of purely signal processing based approaches are evident: To create meaningful transformations the codependencies between different voice properties have to be understood well and modelled precisely. Modelling these correlations with heuristics obtained by empiric studies is not sufficient to create natural results. It is necessary to extract information about the voice systematically and use this information during the transformation process automatically. Recent advances in computer hardware permit this systematic analysis of data by means of machine learning. This thesis thus uses machine learning to create a neural voice transformation framework. The proposed neural voice transformation framework works in two stages: First a neural vocoder allows mapping between a raw audio and a mel-spectrogram representation of voice signals. Secondly, an auto-encoder with information bottleneck allows disentangling various voice properties from the remaining information. The auto-encoder allows changing one voice property while automatically adjusting the remaining voice properties. In the first part of this thesis, we discuss different approaches to neural vocoding and reason why the mel-spectrogram is better suited for neural voice transformations than conventional parametric vocoder spaces. In the second part we discuss the information bottleneck auto-encoder. The auto-encoder creates a latent code that is independent of its conditional input. Using the latent code the synthesizer can perform the transformation by combining the original latent code with a modified parameter curve. We transform the voice using two control parameters: the fundamental frequency and the voice level. Transformation of the fundamental frequency is an objective with a long history. Using the fundamental frequency allows us to compare our approach to existing techniques and study how the auto-encoder models the dependency on other properties in a well known environment. For the voice level, we face the problem that annotations hardly exist. Therefore, first we provide a new estimation technique for voice level in large voice databases, and subsequently use the voice level annotations to train a bottleneck auto-encoder that allows changing the voice level

La veu cantada és probablement l'instrument musical més complex i més ric en matisos expressius. Al llarg de varies dècades s'ha dedicat molt d'esforç a investigar i estudiar les seves propietats acústiques i a entendre els mecanismes involucrats en la producció de veu cantada, posant especial èmfasis en les seves particularitats i comparant-les amb les de la parla. A més, des de l'aparició de les primeres tècniques de síntesi de so, s'ha intentat imitar i sintetitzar per mitjà de tècniques de processament del senyal.

El principal objectiu d'aquesta recerca doctoral és construir un sintetitzador de veu cantada capaç de reproduir la veu d'un cantant determinat, que tingui la seva mateixa expressió i timbre, que soni natural, i que tingui com a entrades només la partitura i la lletra de una cançó. Aquest és un objectiu molt ambiciós, i en aquesta tesi discutim els principals aspectes de la nostra proposta i identifiquem les qüestions que encara queden obertes.
La voz cantada es probablemente el instrumento musical más complejo y el más rico en matices expresivos. A lo largo de varias décadas se ha dedicado mucho esfuerzo de investigación a estudiar sus propiedades acústicas y a entender los mecanismos involucrados en la producción de voz cantada, poniendo especial énfasis en sus particularidades y comparándolas con el habla. Desde la aparición de las primeras técnicas de síntesis de sonido, se ha intentado imitar dichos mecanismos y encontrar maneras de reproducirlos por medio de técnicas de procesado de señal.

El principal objetivo de esta investigación doctoral es construir un sintetizador de voz cantada capaz de reproducir la voz de un cantante determinado, que tenga su misma expresión y timbre, que suene natural, y cuyas entradas sean solamente la partitura y la letra de una canción. Éste es un objetivo muy ambicioso, y en esta tesis discutimos los principales aspectos de nuestra propuesta e identificamos las cuestiones aún sin resolver.
Singing voice is one of the most challenging musical instruments to model and imitate. Along several decades much research has been carried out to understand the mechanisms involved in singing voice production. In addition, from the very beginning of the sound synthesis techniques, singing has been one of the main targets to imitate and synthesize, and a large number of synthesizers have been created with that aim.

The final goal of this thesis is to build a singing voice synthesizer capable of reproducing the voice of a given singer, both in terms of expression and timbre, sounding natural and realistic, and whose inputs would be just the score and the lyrics of a song. This is a very difficult goal, and in this dissertation we discuss the key aspects of our proposed approach and identify the open issues that still need to be tackled.

The sustained increase in computing performance over the last decades has brought enough computing power to perform significant audio processing in affordable personal computers. Following this revolution, we have witnessed a series of improvements in sound transformation techniques and the introduction of numerous digital audio effects to modify effectively the time, pitch, and loudness dimensions of audio signals. Due to the complex and multi-dimensional nature of timbre however, it is significantly more difficult to achieve meaningful and convincing qualitative transformations. The tools currently available for timbre modifications (e.g. equalizers) do not operate along perceptually meaningful axes of singing voice timbre (e.g. breathiness, roughness, etc.) resulting in a transformation control problem. One of the goals of this work is to examine more intuitive procedures to achieve high-fidelity qualitative transformations explicitly controlling certain dimensions of singing voice timbre. Quantitative measurements (i.e. voice timbre descriptors) are introduced and used as high-level controls in an adaptive processing system dependent on the characteristics observed in the input signal.
The transformation methods use a harmonic plus noise representation from which voice timbre descriptors are derived. This higher-level representation, closer to our perception of voice timbre, offers more intuitive controls over timbre transformations. The topics of parametric voice modeling and timbre descriptor computation are first introduced, followed by a study of the acoustical impacts of voice breathiness variations. A timbre transformation system operating specifically on the singing voice quality is then introduced with accompanying software implementations, including an example digital audio effect for the control and modification of the breathiness quality on normal voices.

This book introduces readers to the historical, performative, and cultural context of pansori, a traditional Korean oral story-singing art. Written by a scholar-practitioner of the form, this study is structured in three parts and begins by introducing readers to the technical, aesthetic, and theoretical components of pansori as well as the synthesis of vocal and percussive elements that stage the narrative. It also reflects on the historical contexts of pansori alongside Korea’s transformation from Joseon monarchy to modern statehood. It argues that with colonial annexation came modernist influences that Korean dramatists and audiences used to create distinct new genres of Korean performance, using the common thread of pansori. It further explores the dynamic interplay of preservation and innovation, beginning in the post-war designation of national performance art and continuing with developments that coincide with Korea’s imprint on cultural globalization. Along with Korea’s growth as a world economic center, a growing enthusiasm for Korean culture around the world has increased the transmission and visibility of pansori. Chan E. Park argues that tradition and innovation are not as divergent as they are sometimes imagined to be and that tradition is the force that enables innovation. Unique among treatments of this subject, this book is written from combined researcher and practitioner perspectives. Drawing on her ethnographic work and performance practice, Chan E. Park interweaves expert knowledge of both the textual and performative aspects of the form, rendering legible this dramatic tradition. Pansori as voice theatre examines the practice of pansori, Korean oral storysinging, in its historical, performative, and cultural context. There are three sections to the text. Section One analyzes technical, aesthetic, and theoretical components of pansori. Using translated samples, the book explicates the synthesis of vocal and percussive elements that, together with the listening ear, stage the narrative. Section Two critically reflects on the historical contexts of pansori alongside Korea’s transformation from Joseon monarchy (1391-1910) to modern statehood. In the case of pansori tradition, the posited strict class hierarchies of Joseon were in fact porous, as attested by the patronages and collaborations over generations. With colonial annexation (1910-1945) came modernist influences that Korean dramatists and audiences localized into distinct new genres of Korean performance using the common thread of pansori. Section Three examines the dynamic interplay of preservation and innovation, beginning in the post-War designation of national performance art (gugak) and continuing with off-mainstream, fusion, and hybrid developments that coincide with Korea’s imprint on cultural globalization (hallyu). Along with Korea’s growth as a world economic center (late twentieth century to present), a growing enthusiasm for Korean culture around the world has increased the transmission and visibility of pansori through institutions of popular culture. Ultimately, tradition and innovation are not as divergent as they are sometimes imagined to be. Rather, a study of pansori suggests that tradition is the centripetal force that enables innovation for the collective theatrical imagination.

Chapter 7 discusses gendered political and personal implications of voice loss. It examines the loss of voice through and mediated by technology. First, it traces a history in American cinema and the shifting sociopolitical landscapes in which it has projected its visions and auditions, where the transformation of voices—especially women’s— can be heard clearly. The singing voice in cinema has moved from symbolizing the vulnerability of identity and its susceptibility to manipulation, to embodying the affirmation of identity as a site of individual agency. Outside of cinema, voice loss in singers is subject to discourses no less suffused with ideas about identity and agency. Second, the chapter explores the loss of voice that one singer experienced by selling her voice to a digital sampling library.

San Francisco’s Cockettes troupe staged radical anti-disciplinary spectacles onstage, in short films, in public, and in their domestic spaces. They hyperbolized the leftist political programs that informed both the liberation movements and the communal living practices in which they were ensconced. Through elaborate gender-defying combinations of drag and nudity, the Cockettes used their bodies as sites of social transformation. Sylvester, who later became a recording star, was perhaps the best-known Cockette. Using a repertoire of black virtuoso diva techniques, including a proficient singing voice, attention to black musical forms, and articulate modes of dress, Sylvester perfected a black expressive originality constructed from historically black signs. The difference between his radical virtuosity and the transgressive drop-out aesthetic of the predominantly white Cockettes troupe reveals a lack of organic unity in this revolutionary space and a racial cleavage in the project of liberation.

Abstract This chapter examines community singing in a state prison context, with a focus on Voices of Hope (VOH), a women’s prison choir in Minnesota. After a brief overview of incarceration, music, and gender in US prisons, the author examines the countercultural nature of community singing as it relates to VOH. The author then evaluates three concepts central to the choir’s mission: restorative justice, communal mindset, and awareness/advocacy. Finally, the chapter provides evidence of the transformative power of community singing, using quotes and stories from current and former VOH singers. The author uses a narrative approach, inviting the reader into an empathetic rather than analytical space in an effort to elevate the voices of incarcerated singers.

Abstract In its earliest years opera emerged as an enhancement of festivities designed to glorify the rule of dynasties in city-states or the power of cardinals in Rome. Its arrival in Venice led to a radical transformation in its very essence and a new beginning in its history. Coming from Rome in 1637, what the first opera troupe found upon arriving in Venice was not a dynasty to glorify (unless it was Venice herself) but thriving commerce and, especially during carnival, a teeming international, pleasure-seeking public. Opera took root immediately, on an entirely new basis: in one form or another, the public paid to be admitted to the theater, and the introduction of this new commercial factor speedily had its effect on what the public was offered. Opera very soon learned to adapt itself to the new consumers: scenic effects remained a high priority, but now solo singing grew tremendously in importance. More and more, composers strove to exploit the solo voice in constructing their scores, and star opera singers began to dominate the operatic stage (as they do to this day). It was Venetian opera, in turn, that dominated wherever opera found a new venue, whether in Italy or abroad. And except where supported by kings or other rulers, its economic underpinnings reflected the lessons learned in Venice. The details of operatic production in seventeenth-century Venice are nowhere so clearly described as in the following extracts from a book by the theatrical chronicler Cristoforo lvanovich published in 1681. Entitled Minerva al tavolino (Minerva at her desk), it is a catalogue of all the operas produced at Venice’s numerous theaters from 1637 to the date of publication, with an appendix (from which we quote) describing the theaters themselves and how they functioned.

Over a long career, New Jersey’s Bruce Springsteen became the poet of the working poor, just as Indiana’s John Mellencamp became the voice of embattled farm families and Bob Seger the rock ‘n’ roll bard of the Midwest. Springsteen’s stature became such that, against his own intentions, his appeal was invoked by Republican conservatives during the transformative Reagan era. The success of these artists, singing about the hopes and fears of ordinary citizens whose communities and livelihoods were besieged by globalization, was directly tied to the sweeping shifts of labor and governance that affected the industrialized world in the 1980s.