Concept dictionary
Audio
Audio is the modality that handles sound, including speech, music, noise, and other acoustic signals that AI systems can analyze or generate.
- Review state
- Reviewed in POC
- Freshness
- Current
- Version
- 2
Also called
- Audio
- Audio AI
Remember it this wayAudio is the sound signal; speech is one kind of content inside it.
In plain language
Audio is sound represented so a system can record, transmit, analyze, or generate it. It may contain speech, music, environmental sounds, silence, or noise.
Why it matters
Treating audio as its own modality helps teams account for sampling, background noise, speakers, timing, consent, and accessibility rather than treating sound as text with a different file extension.
Example
A meeting assistant receives an audio stream, separates speakers, transcribes speech, and flags a fire alarm heard in the background.
Common misconception
Audio is not limited to speech. A speech transcript can omit tone, music, noise, timing, and other information present in the original signal.
Related terms
Sources
- W3C, “SMIL 2.0: BasicMedia Module”distinguishes audio, still-image, and video media objects.
- W3C, “Media Fragments Use Cases and Requirements”describes audio and video as time-aligned media tracks.
Learn more
- Compare [[speech-ai|Speech AI]], [[multimodal|Multimodal]], [[video|Video]], and [[natural-language-processing|Natural Language Processing]] to place Audio in the surrounding system.
- Follow the cited source material above for the method, standard, or research context behind this draft.