Audio Analytic

Audio Analytic is a British company headquartered in Cambridge, England that has developed a patented sound recognition software framework called ai3 which provides technology with the ability to understand context through sound. This framework includes an embeddable software platform that can react to a range of sounds such as smoke alarms and carbon monoxide alarms, window breakage, infant crying and dogs barking.

Audio Analytic
Private
IndustrySoftware, Embedded
FoundedCambridge, UK (2010 (2010)) Series A Investment
FounderDr. Christopher Mitchell (CEO)
HeadquartersCambridge, UK
Key people
Dr. Robert Swann (chairman) Alphamosaic, Amy Weatherup (director)
ProductsSound Recognition Systems
Websitewww.audioanalytic.com

History

The company was based on founder Christopher Mitchell's doctoral research from Anglia Ruskin University, with seed investment from EEDA (East of England Development Agency) and local Cambridge Angels investors.

The company's investors include Cambridge Angels, Rockspring, Martlets, Cambridge Investment Capital and IQ Capital Partners. [1]

Development milestones
  • 2011 Audio Analytic announced the version one release of its professional security products Glass Break, Aggression, Car Alarm and Gunshot.
  • 2017 Audio Analytic announced a commercial partnership with hearables company Bragi to embed sound recognition in earphones.
  • 2018 Audio Analytic announced a partnership with Hive (part of Centrica plc) to embed sound recognition technology into multiple products.
  • 2018 Audio Analytic announced that their sound recognition software has been launched alongside Alexa and Netflix on the Freebox Delta from European telecoms company Iliad.

Products

Audio Analytic sells ai3, a software package that is embedded on a device, along with an assortment of sound profiles that the software can recognise, including warning alarms, window breakage, an infant crying, and voice activity. ai3 is hardware-agnostic and can run on a variety of devices, such as smart speakers and smartphones. It is designed to run on-device and not in the cloud.

Audio Analytic developed the Polyphonic Sound Detection Score (PSDS), a metric for evaluating the performance of sound recognition algorithms when applied to polyphonic sound recordings.[2][3][4] They also released an accompanying software framework that implements the PSDS.[5]

gollark: That is an establishment lie.
gollark: Parrots cannot be run rapidly on GPUs.
gollark: Can you *do* that?
gollark: Oh, wait, `gc.get_objects`.
gollark: I may have to rewrite it iteratively.

References
  1. "Audio Analytic - Crunchbase". Crunchbase. Retrieved 4 August 2020.
  2. Bilen, Cagdas; Ferroni, Giacomo; Tuveri, Francesco; Azcarreta, Juan; Krstulovic, Sacha (May 2020). "A Framework for the Robust Evaluation of Sound Event Detection". ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP): 61–65. doi:10.1109/ICASSP40776.2020.9052995.
  3. DCase 2020 Challenges. "Sound event detection and separation in domestic environments - DCASE". dcase.community. Retrieved 4 August 2020.
  4. Wisdom, Scott; Erdogan, Fonseca, Eduardo and Salamon, Justin and Seetharaman, Prem and Hershey, John R., Hakan; Ellis, Daniel P. W.; Serizel, Romain; Turpault, Nicolas; Fonseca, Eduardo; Salamon, Justin; Seetharaman, Prem; Hershey, John R. (2020). "What's All the FUSS About Free Universal Sound Separation Data?". in preparation.CS1 maint: multiple names: authors list (link)
  5. Audio Analytic (22 July 2020). "audioanalytic/psds_eval GitHub repository". GitHub. Audio Analytic. Retrieved 4 August 2020.


This article is issued from Wikipedia. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.