Your data on MRCVSonline
The nature of the services provided by Vision Media means that we might obtain certain information about you.
Please read our Data Protection and Privacy Policy for details.

In addition, (with your consent) some parts of our website may store a 'cookie' in your browser for the purposes of
functionality or performance monitoring.
Click here to manage your settings.
If you would like to forward this story on to a friend, simply fill in the form below and click send.

Your friend's email:
Your email:
Your name:
 
 
Send Cancel

Artificial intelligence could translate dog vocalisation
Researchers adapted a tool previously trained for human speech.

Technology can distinguish between playfulness and aggression.

Researchers from the University of Michigan are exploring how artificial intelligence (AI) could be used to decipher dog barks.

The AI model has the potential to discover information from animal vocalisations, including the dog’s age, breed and sex. The researchers also believe it could identify if a bark is playful or aggressive.

The project saw researchers adapt a speech-processing model, which was previously trained to study human speech.

Through a collaboration with the National Institute of Astrophysics, Optics and Electronics (INAOE) Institute in Mexico, the team discovered that this model could act as a starting point for training new systems for animal communication.

The development of an AI model for animal vocalisations was previously hampered by the lack of public data. Although human samples are easy to record, there are more limitations when collecting animal recordings.

Researchers say that animal vocalisations are logistically more difficult to record as they either need to be recorded in the wild or, for domestic pets, with the permission of owners.

It was due to these limitations that researchers opted to instead repurpose an existing, human-oriented model.

Existing voice technologies, such as voice-to-text and language translation, are trained to identify the nuances of human speech. The tools are able to distinguish between tone, pitch and accent to translate speech and identify speakers.

The team adapted this model by using a dataset of dog vocalisations from 74 different dogs – of varying breed, age, sex and context. These recordings were then used to modify the machine-learning model.

Using this tool, researchers were able to generate and interpret acoustic representations from the dogs. The AI model not only passed four different classification tasks, but also outperformed other models specifically trained on dog barks with accuracy figures of up to 70 per cent.

Rada Mihalcea, from the University of Michigan’s AI laboratory, said: "This is the first time that techniques optimised for human speech have been built upon to help with the decoding of animal communication.

"Our results show that the sounds and patterns derived from human speech can serve as a foundation for analysing and understanding the acoustic patterns of other sounds, such as animal vocalisations."

The full study can be found here.

Image © Shutterstock

Become a member or log in to add this story to your CPD history

Antibiotic Amnesty survey opens

News Story 1
 The Responsible Use of Medicines Alliance - Companion Animal and Equine (RUMA - CA&E) has launched a survey which seeks to measure the success of this year's Antibiotic Amnesty campaign.

Its results will help the group to calculate how many unused or out-of-date antibiotics were returned as part of the campaign throughout November. Practices are also asked how they engaged with clients to raise awareness of the issue.

Participating practices can enter a prize draw to win up to 100 in vouchers for taking part. Further prizes from BSAVA and Webinar Vet will be available for practices with the best input.

The survey takes a few minutes to complete and will be running throughout December. It can be accessed via this link

Click here for more...
News Shorts
Bluetongue cases lead to restricted zone extension

Following the detection of new cases of bluetongue virus serotype 3 (BTV-3), the restricted zone has been extended to include additional parts of Hampshire, Wiltshire, Oxfordshire and North Yorkshire, as well as Bournemouth, Christchurch, Poole and part of Dorset.

The zone now covers 28 counties and unitary authorities in England. As of 16 December, 185 cases of BTV-3 have been confirmed in England and Wales during the current vector period.

Bluetongue is a notifiable disease. Suspected cases must be reported on 03000 200 301 in England or 03003 038 268 in Wales. In Scotland, suspected cases should be reported to the local field services office. Suspected cases in Northern Ireland should be reported to the DAERA Helpline on 0300 200 7840 or by contacting the local DAERA Direct Veterinary Office.

A map of the area where movement restrictions apply can be found here.