Linguitone: A Web-based System to Assist in Making Electric Guitar Sounds Using Natural Language Expressions
Guitarists often use unique linguistic expressions to convey the image of sound to other guitarists. However, beginners find it difficult to relate those expressions to the actual sound image. It is also challenging for them to realize the sound they have in mind with effect pedals.
In this study, we focused on linguistic expressions describing timbres. We developed Linguitone, a system that supports sound design using a machine learning model that predicts parameters of guitar effectors from embedding vectors of natural language. In addition, we evaluated the usability and supportability of the system with electric guitar players.