ASEAN's Premier News and Eco System Portal for Big Data, IoT and AI

Home > Big news > News > SCDF turns to artificial intelligence to help emergency call dispatchers
SCDF turns to artificial intelligence to help emergency call dispatchers

 

With Singapore’s emergency dispatch phone operators receiving almost 200,000 calls for assistance a year, every minute is vital.

In an effort to ease their workload, the Singapore Civil Defence Force (SCDF) and four other Government agencies are turning to artificial intelligence (AI), using a speech recognition system that can transcribe and log each call received in real time – even if it is in Singlish.

For now the system is programmed to recognise English and Mandarin with some Hokkien and Malay, though it could be customised to incorporate others.

AI Singapore (AISG), a programme under the National Research Foundation, is investing $1.25 million to set up the AI Speech Lab, which is headed by the two professors who created the system.

It was developed using artificial intelligence, such as deep learning technology, which works off algorithms that mimic the human brain’s neural pathways to help computers perform new tasks and analyse data.

“If successful, it will improve how SCDF’s emergency medical resources are dispatched and enhance the overall health outcomes of those in need,” said the SCDF’s director of operations, Assistant Commissioner Daniel Seet.

He added that it will help to reduce the time it takes the SCDF’s 995 operations centre dispatchers to log in information.

Dispatchers ask the caller questions to determine the nature and severity of the case, to make sure the appropriate emergency medical resources are sent.

“In an emergency, every minute counts,” he added.

The AI Speech Lab is led by Professor Li Haizhou, an expert in speech, text and natural language processing from the National University of Singapore, and Associate Professor Chng Eng Siong from the Nanyang Technological University.

They have been working on the speech recognition system for around a decade.

Prof Li said such a code-switching, or mixed-lingual, system is currently not commercially available.

He said: “This technology performs better than commercial engines as it can accurately recognise conversations comprising words from different languages. It solves a unique Singapore problem.”

To develop the system, researchers collected over 1,000 hours of speaking samples from Singapore and Penang – a state that mixes languages in speech similar to Singapore – as well as recordings of Singaporeans from radio stations, YouTube and SoundCloud.

These recordings are manually transcribed into text. The system then “learns” the association between the text and the collected speech samples.

The system has “learnt” about 40,000 English and Mandarin words each, and has an accuracy rate of about 90 per cent.

Unique words the system can recognise include “jiak ba bueh” and “hoh boh” – “have you eaten” and “how are you” in Hokkien – and local dishes such as char kway teow and nasi lemak.

The lab is staffed by five AI engineers and located at the innovation 4.0 building on NUS’s Kent Ridge campus.

Professor Ho Teck Hua, executive chairman of AISG, said the system could also benefit companies, as it can be customised according to their business needs.

Mr Tan Kok Yam, deputy secretary of the Smart Nation and Digital Government agency, said: “The Government is keen to harness artificial intelligence to serve our citizens better. GovTech is collaborating with AISG to develop solutions that can improve planning and service delivery.”

Research director at research and advisory company Gartner, Mr Manjunath Bhat, said: “Multi-lingual speech transcription will make it easy for senior citizens and people speaking all dialects to participate in digital initiatives.

“Even as communication systems switch from analogue to digital, human language itself remains analog. The new solution enables computers to speak in the language of the common person as opposed to humans learning to adapt to digital interfaces.”

(0)(0)