Selected audio recordings

2011

in: Dissertation Abstracts International A 75,7(E), Jan. 2015. This research focuses on singable translating in the Cantonese dubbed version of Disney animated musicals, one of the most complicated – but at the same time one of the most neglected – translational activities. The present research is a pioneer attempt at an in-depth study of the relationship between the target text and four major components of the multimedial presentation respectively, namely, the source music (one-melody relationship, the time-related elements, and rhyming), the source text (semantic fidelity, semantic deviation and semantic anomaly), the source images (word-image relationship), and the target viewer (immediate comprehensibility), hoping to shed new light on this complicated yet neglected translational activity. A viewer-oriented approach for translating dubbed songs into Cantonese is identified. Cheng, Xingwang: Zǎoqí zhōngguó dònghuà diànyǐng yīnyuè jíqí lìshǐ dìwèi. In: Zhōngyāng Yīnyuè Xuéyuàn xué...

In questo trattato si pone l’obiettivo di analizzare la gestione musicale nei cartoni animati della “Golden Age” (anni ’40), prendendo in esame i compositori Carl Stalling e Scott Bradley, studiando la loro metodologia compositiva, il contesto lavorativo in cui erano immersi e le relative aspirazioni e ideali che hanno influito sulla forma mentis delle loro composizioni.

Glasba in/za oder / Music in/forStage, The 28th Slovenian Music Days, (March 12-14 2013), Ljubljana (Slovenija), 2014

2019

This thesis tries to elucidate and exemplify possible intersections between experimental rock of the late 60s and early 70s and the field of electroacoustic music, focusing mostly on production/compositional techniques and aesthetic approaches. A historical overview of the context which experimental rock emerged from is attempted, exploring why and how certain production techniques were used at that period in rock music and investigating into whether the aesthetic outcome of these techniques relates to experiments in the field of electroacoustic music. Musical examples from the field of classical, jazz and rock styles (including artists like Glenn Gould, Lennie Tristano, Miles Davis, the Beatles, Pink Floyd, Fred Frith (Henry Cow) and Frank Zappa) are addressed for the sake of exploring different experimental studio practices. Moreover, a part of this research deals with the creative exploration of the production techniques discussed in order to develop my personal compositional work and eventually to create a framework where a discussion about composing electroacoustic music from the perspective of rock production (and vice-versa!) could emerge.

The history of jazz is one of cascading events and styles that continually inform and influence what follows. This transformation is an accumulation over time of styles, methods, and techniques. During the 1950s and 1960s, electronic music gradually left the confines of the institution and migrated into the hands of independent musicians. Among the early explorers were jazz musicians who laid a foundation of experiment for later musicians to model and emulate. This article discusses two types of such experiments: (1) Jazz incorporating prerecorded electronic music on tape (1960–1970) and (2) Jazz using live electronics other than synthesizers (early years from 1965–1970). A discography is provided.

Natural Language Processing Journal Volume 7, 2024

Natural language processing and vision tasks have recently seen large improvements through the rise of Transformer architectures. The high-performing large language models (LLMs) benefit from large textual datasets that are numerously available online. However, action and bidirectional action-language tasks are less developed, as these require more specific and labeled data. Therefore, we aim at enabling these robotic action capabilities for a pretrained LLM, while maintaining high efficiency with regards to the required training time and data size. To achieve this, we split up a Transformer-based LLM and insert a multimodal architecture into it. Specifically, we split a pretrained T5 LLM between its encoder and decoder parts, to insert a crossmodal Transformer component of a Paired Transformed Autoencoders (PTAE) bidirectional action-language model. The experiments are conducted on a new dataset, consisting of unimodal language translation and crossmodal bidirectional action-language translation. The natural language capabilities of the original T5 are re-established efficiently by training the crossmodal Transformer, which requires only one 5.7 millionth of the T5 model's original training data. Furthermore, the new model, called CrossT5, achieves high accuracy for the vision-and language-guided robotic action tasks. By design, the CrossT5 agent acts robustly when tested with language commands not included in the dataset. The results demonstrate that this novel approach is successful in combining the advanced linguistic capabilities of LLMs with the low-level robotic control skills of vision-action models. The code is available at this URL: https://github.com/samsoneko/CrossT5.

Anais do XIII Seminário Internacional de Políticas Culturais , 2024

RESUMO: O presente artigo discute a importância de entender a cultura como direito e como trabalho a partir da experiência da disciplina Ética e Direitos Culturais, oferecida como componente curricular obrigatório na Faculdade de Comunicação da Universidade Federal da Bahia e ministrada a partir de 2023 para o Curso de Graduação em Produção e Comunicação em Cultura. Pensando a formação em Produção Cultural e as(os) trabalhadoras(es) da cultura no contexto de Salvador-Bahia, defendemos o quão fundamental é a consciência de gênero, raça e classe não só para refletir sobre as condições de trabalho na área, como também transformar tal realidade, inspiradas pela ética de ensinar para transgredir de bell hooks.

Tan razonable como representar una prisión de cierto género por otra diferente es representar algo qué existe realmente por algo que no existe.