Pre-Publication V3. A generative AI process designed to synthesize audio effects elements (e.g. non-musical sound effects, ambient soundscapes, and environmental audio) from text descriptions. or scene context. Distinct from music generation or voice synthesis, these systems are trained to focus on sound effects s and Foley libraries to produce like audio assets.
Deliberation Summary:
(1) Specify and qualify “sound effects” in term, since a general text-to-audio term would have to include music and voice. (2) Review clarification between live recorded effects work and the resulting sound effects, and how far the entry should reach into environmental audio and musical sound design, favoring an open-ended list of audio effect elements over rigid qualification. (3) Note that current systems generate sound not only from direct text prompts but also from inferred scene context and accompanying data.
Pre-Publication V2. The application of generative AIA generative AI process designed to synthesize audio elements, including but not limited to, non-musical sound effects, ambient soundscapes, and environmental audio from text descriptions or scene context. Distinct from music generation and voice synthesis, these systems are trained on sound effects and Foley libraries to produce contextually appropriate like audio assets for scenes or environments not practically recorded.
Deliberation Summary:
Consider grouping entries by workflow type, since several of them are variants of the same input-to-output relationships and wording clarification.
Pre-Publication V1. The application of generative AI to synthesize non-musical sound effects, ambient soundscapes, and environmental audio from text descriptions or scene context. Distinct from music generation and voice synthesis, these systems are trained on sound effects and Foley libraries to produce contextually appropriate audio assets for scenes or environments not practically recorded.