Northumbria switches to speech recognition
- 17 March 2005
A trust-wide implementation of advanced digital dictation and speech recognition software is planned for Northumbria Healthcare NHS Trust. The aim will be to reduce turnaround time and transcription blockages in the trust’s one million items of patient correspondence every year.
Piloting of the new system, G2 Speech’s SpeechReport with Philips’ SpeechMagic, will start in May and the plan is to make a full transfer over from the old analog dictation system in the next 12 months.
Acting general manager, Martin Wilson, said medical staff had been very interested and positive. “We’ve had no shortage of volunteers for the pilot. It is oversubscribed three times.”
He explained that consultant staff and some nurses worked in different hospitals across the trust’s vast, mostly rural area stretching north from the Tyne to the Scottish border. “It’s not really prudent to have different administrative systems across different hospitals.”
The package from Philips and G2 Speech offers options designed to suit users who wish to stay with their current work routines and those who want to switch to full speech recognition. To use full speech recognition, users have to ‘train’ their computer to recognise their voice over a few weeks. Their dictated voice files are then transcribed digitally and sent back to the clinician for checking.
For users who do not want to convert to full speech recognition, Wilson explained that two options are available. The first is digital dictation in which the user dictates correspondence for typing by a secretary. This is the process closest to conventional dictation but users add in extra information including patient demographics from a barcode, an alert for urgent correspondence and a note of what type of communication the dictated file contains. The dictation is also improved by better sound quality.
Under the second option, called batch speech recognition, the dictated sound file is transcribed digitally, but a secretary corrects the document produced by speech recognition before sending it back to the user.
Northumbria has 1500 users in nine locations. The new system will be available to all medical specialties. It is thought to be first time in the UK that speech recognition has been rolled out on such a wide scale, though the technology is used at other centres, notably in radiology.