Back to results
US 2014/0161270 A1Published application
Room identification using acoustic features in a recording
AssigneeInternational Computer Science Institute
InventorsNils PETERS, Howard Lei, Gerald Friedland
Life of this patent
Estimating the term…
Abstract
Analysis of the audio component in multimedia data is disclosed. Rooms can be described through room impulse responses (RIR), the “fingerprint” of a specific room. The method uses machine learning techniques to identify rooms from ordinary audio recordings.
Claims18
A method for measuring reverberation properties of a room comprising the steps of:
- obtaining audio data for a room of a known type;
- analyzing the audio data to extract audio features;
- using machine learning to create a model that describes specific acoustic aspects of the room based upon the audio features;
- and
- storing the model in a database.
2 more elements
A method of identifying a room comprising the steps of:
- storing a database of a plurality of acoustic models for a plurality of identified rooms;
- inputting audio data from at least one microphone in an unknown room;
- extracting audio features from the input audio data;
- using machine learning to create an acoustic model of the unknown room based upon the extracted audio features;
- comparing the acoustic model of the unknown room to the plurality of acoustic models in the database;
- determining if the acoustic model of the unknown room is within a predetermined threshold of at least one of the acoustic models in the database;
- identifying the acoustic model of the unknown room as similar to the room in the database;
- and
- output the identity of the unknown room.
6 more elements
Dependencies are inferred from the claim text.