A Web Audio Node for the Fast Creation of Natural Language Interfaces for Audio Production

Michael Donovan, Prem Seetharaman, Bryan A Pardo

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Audio production involves the use of tools such as reverberators, compressors, and equalizers to transform raw audio into a state ready for public consumption. These tools are in wide use by both musicians and expert audio engineers for this purpose. The typical interfaces for these tools use low-level signal parameters as controls for the audio effect. These signal parameters often have unintuitive names such as “feedback” or “low-high” that have little meaning to many people. This makes them diffi cult to use and learn for many people. Such low-level interfaces are also common throughout audio production interfaces using the Web Audio API. Recent work in bridging the semantic gap between verbal descriptions of audio effects (e.g. “underwater”, “warm”, “bright”) and low-level signal parameters has resulted in provably better interfaces for a population of laypeople. In that work, a vocabulary of hundreds of descriptive terms was crowdsourced, along with their mappings to audio effects settings for reverberation and equalization. In this paper, we present a Web Audio node that lets web developers leverage this vocabulary to easily create web-based audio effects tools that use natural language interfaces. Our Web Audio node and additional documentation can be accessed at https://interactiveaudiolab.github.io/audealize_api.
Original languageEnglish (US)
Title of host publicationProceedings of 3rd Web Audio Conference
PublisherQueen Mary University of London
Number of pages4
StatePublished - 2017

Fingerprint Dive into the research topics of 'A Web Audio Node for the Fast Creation of Natural Language Interfaces for Audio Production'. Together they form a unique fingerprint.

  • Cite this

    Donovan, M., Seetharaman, P., & Pardo, B. A. (2017). A Web Audio Node for the Fast Creation of Natural Language Interfaces for Audio Production. In Proceedings of 3rd Web Audio Conference Queen Mary University of London.