> For the complete documentation index, see [llms.txt](https://www.mlcompendium.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://www.mlcompendium.com/predictive-ml/audio-terminology.md). # Audio Terminology The algorithm pages name their tasks as if the reader already knows them, so the terms come first. The page defines audio source separation, then sound event detection, then query-based separation, which narrows separation down to one sound you ask for. The same notes are in [Audio Source Separation](/predictive-ml/audio-algorithms.md#audio-source-separation) and [Sound Event Detection](/predictive-ml/audio-algorithms.md#sound-event-detection). Audio source separation is the task of pulling specific sound sources out of an audio mixture. Source separation techniques in audio processing can be classified into various methods, such as blind source separation, supervised source separation, and semi-supervised source separation, and these techniques use algorithms and signal processing methods to extract specific sound sources from an audio mixture. Note that usually these methods are for multi-channel audio; if we have a single-channel audio recording, this is more challenging. Separation asks what the sources are; sound event detection asks when they happen. Sound Event Detection (SED) is the task of recognizing the sound events and their respective temporal start and end time in a recording. Sound events in real life do not always occur in isolation, but tend to considerably overlap with each other, and recognizing such overlapping sound events is referred to as polyphonic SED. Query-based separation combines the two ideas by targeting one source. Query-based separation in audio could involve using specific sounds or queries to identify and extract particular sounds from an audio recording. For example, in a crowded audio environment, you might use query-based separation to extract the voice of a specific speaker from a mixture of voices. By providing a query related to the speaker's voice characteristics or specific phrases they are saying, algorithms can be designed to identify and separate that particular speaker's voice from the overall audio recording. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://www.mlcompendium.com/predictive-ml/audio-terminology.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.