Brüel & Kjær Sound & Vibration Measurement A/S - the world’s leading supplier of advanced technology for sound and vibration - puts focus on the importance of realistic testing when developing and incorporating smart speaker functionality into consumer products.
Image Credit: HBK – Hottinger, Brüel & Kjær
An often-used method when testing speech intelligibility and sound quality in voice-operated devices is to measure the response accuracy rate (RAR), which is done by replaying recorded voice commands and evaluating how often the voice command is correctly perceived and responded to. A common approach is to use a standard loudspeaker and microphone, but this can give a false indication of the performance because it doesn’t accurately reproduce the directivity and frequency response of a human voice. The precise method is to use mouth and ear simulators.
“Fortunately, reproducing human speech is not a new challenge,” says Lars Birger Nielsen, Focus Market Manager Brüel & Kjær. “Acoustic engineers - and Brüel & Kjær - have wrestled with the issue for many years in the development of traditional communication devices and have developed standardised test methods for evaluating human communication.”
Brüel & Kjær offers market-leading products that facilitate more realistic speech synthesis and listening by faithfully reproducing human characteristics with very tight tolerances on accuracy to ensure repeatability. Its mouth and ear simulators - led by the new High-frequency Head and Torso Simulator (HATS) – accurately match human voice characteristics, thus ensuring quality and efficiency in smart device testing and development.
The compay’s High-frequency HATS Type 5218 family is the new standard in the field of product audio evaluation. Its capability of both issuing voice commands and measuring the quality of the smart speaker response makes fully automated testing of smart speakers and other voice operated devices possible. Brüel & Kjær also offers a Mouth Simulator Type 4227; a high-performance artificial mouth, which simulates human speech dispersion patterns.