Oxford University Hospitals (OUH) NHS Foundation Trust has shared outcomes from its pilot of ambient voice technology (AVT), noting it shows “real promise” in helping clinicians save time, reducing stress, and allowing more of a focus on patient care.
Taking to LinkedIn, the trust highlights that almost 90 percent of clinicians reported reduced documentation time, also noting that feedback mentioned improved wellbeing and reduced administrative burden. 99.7 percent of patients had consented to the use of AVT, and “found the process efficient and helpful”, it states.
The pilot involved the use of four different AVT solutions: Accurx Scribe, Tortus, Heidi, and Lyrebird, in a range of inpatient and outpatient settings, which the trust reports offered “differently levels of complexity and workflow needs”. From there, an evaluation report was published offering insights into the potential for future adoption.
In the evaluation report, the trust states AVT tools were “generally well received” by clinicians, with 88 percent experiencing measurable time savings, and many describing “a transformation in their working day, as administrative tasks that previously extended into evenings and weekends were now routinely completed within clinic hours”.
47.3 percent of clinicians estimated a saving of 1-10 minutes per encounter; with 45.5 percent estimating this as 5-15 minutes; and 22.2 percent reporting saving an hour or more per day. The percentage of users spending 5-10 hours of their own time on documentation fell from 24.2 percent to 2.2 percent. “This shift led to a reported improvement in work-life balance, reduced cognitive fatigue, and less stress, with staff repeatedly highlighting the positive impact on their wellbeing of these factors being reduced,” the trust notes.
Patient feedback was also “overwhelmingly positive”, according to OUH, with feedback showing patients felt their consultations were more focused and that documentation was thorough and timely.
Although the tools produced the bulk of documentation, reducing effort required to create it from scratch; OUH says “almost all outputs required some degree of editing”, including for minor transcription errors, clarification of terminology, and “occasional inaccuracies”. In total, 37.3 percent of logged AVT outputs required some editing for accuracy. Of 398 uses, 126 edits were needed for accuracy of clinical information, 43 for spelling or grammar, and 24 for hallucinations.
“When giving summative feedback in post-pilot surveys, hallucinations had been experienced by 44.4 percent of users in one or more of their encounters across the range of providers and use cases – illustrating that checking of notes by clinicians is an important process step,” the trust reports. Examples given of hallucinations involved added advice that was not given, the addition of another person in the room who had not been there, and the invention of growth measurements and centiles. “These more significant hallucinations were most commonly observed in complex, multi-voice, or less structured clinical settings, where the technology struggled to capture the nuance and detail required for high-quality clinical records,” it continues.
In terms of other findings, OUH states that there was little variation between the different AVT solutions, that the need to copy and paste outputs was considered “acceptable” given the benefits offered, and that benefits were most pronounced in workflows which best suited AVT.
The trust’s recommendations to maximise benefits are to balance standardisation in documentation with the flexibility to tailor templates to local needs, use clear information around data governance to build trust with clinicians, and adopt a phased approach to deployment. A single solution should serve “a wide range” of use cases, with the addition of a small number of licences for more specialised solutions in complex areas, it continues.
Megan Morys-Carter, director of digital innovation at OUH, highlighted that the initiative is “one of the most comprehensive evaluations of AVT undertaken in an acute NHS setting to date”, adding: “This successful pilot shows how AVT can make a real difference to clinicians’ day‑to‑day work. Staff told us it saves them time, reduces stress, and helps them focus more on patients. These findings give us confidence that AVT has real potential to support future digital transformation at OUH, and across other acute settings in the NHS. It also gave us valuable insights into how to structure procurement and implementation to ensure benefits are realised for hospitals and their staff and patients.”
To read the evaluation report in full, please click here.
Wider trend: Ambient Voice
HTN was joined by a panel of experts from across the health sector for a focused webinar on the use of ambient scribe technology in NHS trusts. Panellists included Lauren Riddle, transformation programme manager at Hampshire and Isle of Wight Healthcare (HIoW); Ynez Symonds, CNIO at HIoW; Dom Pimenta, co-founder and CEO at Tortus AI; and Stuart Kyle, consultant rheumatologist and clinical lead for outpatient transformation at Royal Devon University Hospital. Our panel discussed the practicalities and considerations for ambient scribe implementations, from operating procedures and policies, integration and functionality, through to best practices around patient-practitioner interactions.
NHS England’s Transformation Directorate has published a list of 19 suppliers who have evidenced the criteria required to be part of its self-certified Ambient Voice Technology (AVT) registry – “a national capability to support safe and effective scaling and adoption of AVT across the health and care system”. The 19 suppliers include 33n, Accurx, Anathem, Aprobrium (Lexacom), Beam Up, Corti, Dictate IT, eConsult, HealthOrbit AI, Heidi Health, Lyrebird Health, Microsoft Dragon, Optum (EMIS), Pungo t/a Joy, Scribetech, Tandem, Tortus, T-Pro, and X-On Health.
South East Coast Ambulance Service (SECAmb) has launched a pilot of Tortus’s Ambient Voice solution in its emergency operations centres, looking to free-up clinician time to assist more patients. The tool listens in on conversations, transcribing spoken words into structured medical notes to be checked and approved. According to the trust, this is expected to reduce the amount of time taken by clinicians writing up notes following calls. Alongside the SECAmb pilot, due to run to February 2026, other ambulance services are also committed to testing the technology, including London Ambulance Service and South Western Ambulance Service.
