Active learning of intuitive control knobs for synthesizers using gaussian processes

Huang, Cheng-Zhi Anna; Duvenaud, David; Arnold, Kenneth C.; Partridge, Brenton; Oberholtzer, Josiah W.; Gajos, Krzysztof Z.

doi:10.1145/2557500.2557544

Cited by 9 publications

(9 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To learn nonlinear relationships, one would have to develop an alternative method, although the underlying interaction paradigm of evaluation of examples would remain the same. There has been some follow-on work that explores how to learn nonlinear relationships from user feedback on setting synthesizer parameters (Huang et al 2014), so we believe this approach is generalizable. In this article, however, we advocate for the interaction paradigm rather than arguing in favor of a particular algorithm used to learn from this interaction paradigm.…”

Section: Socialeq: An Evaluative Interfacementioning

confidence: 99%

Learning to Build Natural Audio Production Interfaces

Pardo

Cartwright

Seetharaman

et al. 2019

Arts

View full text Add to dashboard Cite

Improving audio production tools provides a great opportunity for meaningful enhancement of creative activities due to the disconnect between existing tools and the conceptual frameworks within which many people work. In our work, we focus on bridging the gap between the intentions of both amateur and professional musicians and the audio manipulation tools available through software. Rather than force nonintuitive interactions, or remove control altogether, we reframe the controls to work within the interaction paradigms identified by research done on how audio engineers and musicians communicate auditory concepts to each other: evaluative feedback, natural language, vocal imitation, and exploration. In this article, we provide an overview of our research on building audio production tools, such as mixers and equalizers, to support these kinds of interactions. We describe the learning algorithms, design approaches, and software that support these interaction paradigms in the context of music and audio production. We also discuss the strengths and weaknesses of the interaction approach we describe in comparison with existing control paradigms.

show abstract

Section: Socialeq: An Evaluative Interfacementioning

confidence: 99%

Learning to Build Natural Audio Production Interfaces

Pardo

Cartwright

Seetharaman

et al. 2019

Arts

View full text Add to dashboard Cite

show abstract

“…These deeper insights have been shown to facilitate more effective recombination of ideas than with raw example ideas [28,54]. These maps can also improve iteration on ideas by enabling people to discover and explore many closely related solution alternatives [9,17,27,38].…”

Section: Creativity Enhancing Interventionsmentioning

confidence: 99%

“…Human computation approaches have been successfully applied to organize artifacts in various domains such as 3D modeling [9,47], graphic designs [37,38] and music composition [17]. These approaches all require considerable number of inputs from humans to discover the design space of ideas; some of these inputs are extracted from users' interactions with the system [17,47], but most of these inputs are from small, explicit human computation tasks, such as clustering subsets of items [1,10], completing similarity comparisons between items [44], or identifying attributes of items [9,37,38].…”

Section: Uncovering Semantic Relationships In Large Corporamentioning

confidence: 99%

IdeaHound

Siangliulue

Chan

Dow

et al. 2016

Proceedings of the 29th Annual Symposium on User Interface Software and Technology

Self Cite

View full text Add to dashboard Cite

Prior work on creativity support tools demonstrates how a computational semantic model of a solution space can enable interventions that substantially improve the number, quality and diversity of ideas. However, automated semantic modeling often falls short when people contribute short text snippets or sketches. Innovation platforms can employ humans to provide semantic judgments to construct a semantic model, but this relies on external workers completing a large number of tedious micro tasks. This requirement threatens both accuracy (external workers may lack expertise and context to make accurate semantic judgments) and scalability (external workers are costly). In this paper, we introduce IDEAHOUND, an ideation system that seamlessly integrates the task of defining semantic relationships among ideas into the primary task of idea generation. The system combines implicit human actions with machine learning to create a computational semantic model of the emerging solution space. The integrated nature of these judgments allows IDEAHOUND to leverage the expertise and efforts of participants who are already motivated to contribute to idea generation, overcoming the issues of scalability inherent to existing approaches. Our results show that participants were equally willing to use (and just as productive using) IDEAHOUND compared to a conventional platform that did not require organizing ideas. Our integrated crowdsourcing approach also creates a more accurate semantic model than an existing crowdsourced approach (performed by external crowds). We demonstrate how this model enables helpful creative interventions: providing diverse inspirational examples, providing similar ideas for a given idea and providing a visual overview of the solution space.

show abstract

“…Huang [15] made a control knob for synthesis of new sounds parameterized by a natural language word (e.g. "scary"), learned from interaction with a user.…”

Section: Content Creation/control With High-level Descriptorsmentioning

confidence: 99%

Audealize: Crowdsourced Audio Production Tools

Seetharaman¹,

Pardo²

2016

J. Audio Eng. Soc.

View full text Add to dashboard Cite

Audio production involves using tools such as reverberators and equalizers to transform audio into a state ready for release. While these tools are widely available to musicians who are not expert audio engineers, existing interfaces can be frustrating for newcomers, as their interfaces are parameterized in terms of low-level signal manipulations that may not be intuitive to non-experts. We present Audealize, an interface that bridges the gap between low-level parameters of existing audio production tools and programmatic goals (e.g. "make my guitar sound 'underwater"'). Users modify the audio by selecting descriptive terms in a word-map built from a crowdsourced vocabulary of word labels for audio effects. We perform the first user study comparing a word-map interface to traditional audio production interfaces. A study on a population of 432 non-experts found they favored the crowdsourced word-map over traditional interfaces. Absolute performance measures show those using the word-map interface produced results that equaled or exceeded results using traditional interfaces. This indicates language, in concert with a meaningful word-map visualization, is an effective interaction paradigm for audio production by non-experts. Audealize is an example of a general design approach that can apply to other domains and can be accessed at http://audealize.appspot.com.

show abstract

Active learning of intuitive control knobs for synthesizers using gaussian processes

Cited by 9 publications

References 14 publications

Learning to Build Natural Audio Production Interfaces

Learning to Build Natural Audio Production Interfaces

IdeaHound

Audealize: Crowdsourced Audio Production Tools

Contact Info

Product

Resources

About