How Singapore is using AI to improve emergency services
Have you ever asked Siri or Alexa to “text mom ‘I love you’, only to find that the message says “rainbow blue” instead? Voice assistants can be extremely helpful, but our messages sometimes get lost in translation.
Such is the fallibility of speech recognition. When an individual speaks in multiple languages, these programs may not be able to understand them at all. This is why AI Singapore has developed a voice recognition program capable of recognizing colloquial English spoken in the country.
Laurence Liew, Director of AI Singapore (AISG), explains how the program helps emergency services and explains how governments can put AI to good use.
AI for emergency services
Singapore established ISGF to nurture local AI talent and spearhead AI innovation to create tools for good. The organization helped the Singapore Civil Defense Force (SCDF) dispatch emergency medical resources faster through an emergency call transcription program.
This program can reduce the time it takes for dispatchers to jot down information from the call, such as the severity of any injuries, the Straits Times reported. This will improve how the SCDF distributes its emergency medical resources, Deputy Commissioner Daniel Seet, the Force’s director of operations, said in the same report.
Although there are existing transcription programs, many of them only work for those who speak standard English, shares Liew. In Singapore, however, the norm is to speak a mix of languages, including Singlish – a form of colloquial English that incorporates multiple languages like English, Mandarin and Malay.
Many of these programs won’t work when multiple languages or Singlish are taken into account, Liew says. This ISGF program is able to transcribe emergency calls in Singapore with an accuracy rate of around 90%, the Straits Times wrote.
He is also able to recognize Hokkien terms such as “jiak ba bueh”, which means “have you eaten?” and even the names of local dishes like “char kuey teow,” the article says.
This program also allows SCDF to better plan its longer-term emergency services resources, Liew shares. The Mounted Police can perform text and data analysis on transcripts to identify times and areas where emergency services are most needed.
For example, the SCDF parked ambulances at local community centers during morning and evening rush hours after analyzing data from their past emergency responses, according to a Facebook post on their page.
Such data analysis would be difficult without accurate transcripts, Liew adds.
AI in government
Beyond emergency services, AI has immense potential to help governments deliver better services to citizens. Countries around the world use it to fight climate change, improve safety, or even diagnose diseases in their plants.
AI will automate, streamline, improve productivity and reduce the cost of delivering services to citizens. But this is often accompanied by fears of cutting millions of jobs.
It’s not true. Instead, AI often automates tasks workers don’t want to do because they’re “boring, mundane, or dangerous,” Liew shares. This frees up the workforce to focus on other more important tasks, he explains.
For example, when email became popular, people often had to spend the first 15 minutes of their workday cleaning their inboxes of spam. Email today is automated so spam is automatically detected, tagged and removed from the inbox, he shares.
Rethinking Talent Hiring and Retention in AI
With AI being such a competitive industry, government agencies need to think about new ways to hire and retain AI talent if they want to grow their AI capabilities, suggests Liew.
Indeed, talent experienced in AI is often sought after by large tech companies. It is difficult for governments to compete with these companies because the pay scale in governments is often fixed, he explains.
ISGF solves this problem through its apprenticeship program, which brings in AI graduates and trains them for nine months. Thereafter, many of their apprentices stay on to complete a two-year contract before leaving. At this point, a new wave of apprentices will join the team.
This is a new pattern that HR may not be familiar with, but which is inevitable in the AI and machine learning talent wars, Liew shares. With this model, ISGF has managed to grow from a team of just four engineers to almost 40 people today, he adds.
With AI, governments can divert their attention from menial tasks like email and focus instead on delivering better services both in emergencies and in the daily lives of citizens. All it takes is a spark of innovation, both in technology and in hiring.
Want to learn more about how governments can use AI? Laurence Liew will speak at AI x GOV as a panelist on “The Future of AI in Government”, April 6, 4:45-5:45 GMT+8. Register here!