Health Consultations

A Healthcare System with a Language Problem It Couldn't Solve Alone

South Western Sydney is home to one of Australia's most culturally diverse populations. In suburbs like Fairfield and Bankstown, more than half of residents speak a language other than English at home, including Arabic, Vietnamese, Mandarin, Cantonese, and many others.

For the clinicians at Fairfield Hospital and Bankstown-Lidcombe Hospital, this creates a daily reality. A midwife navigating a complex antenatal consultation with a patient who speaks only Arabic is not just facing a communication gap; she's managing clinical risk.

The demand for medical interpretation had outpaced what traditional interpreter services could provide. SWSLHD needed something fundamentally different, not a patch, but a platform.

The Barriers SWSLHD Faced Every Day
The Barriers SWSLHD Faced Every Day

Why VideoTranslatorAI?

In 2022, the NSW Government's Small Business Innovation & Research (SBIR) Program, run by the Office of the NSW Chief Scientist & Engineer, issued the Cultural and Linguistic Diversity Challenge in partnership with NSW Health and SWSLHD. The brief: find an AI-powered solution capable of genuinely improving healthcare communication for CALD communities at scale.

VideoTranslatorAI's proposal was selected from a competitive pool. The decision reflected not just technical capability, but a rigorous methodology: the team proposed real audit, reporting, and remediation processes to ensure AI accuracy in complex medical conversations, something previous attempts with generic translation tools had failed to deliver.

The SBIR Journey

Phase 1 — Feasibility Study

Phase 1 focused on a foundational question: could AI-driven voice-to-text transcription and translation be made accurate, auditable, and safe enough for real clinical use? The study concentrated on antenatal visits, one of the most linguistically and emotionally complex consultation settings in medicine.

The VideoTranslatorAI team worked directly on the ground with clinical leaders, including Professor Josephine Chow and Clinical Midwifery Consultants at Fairfield and Bankstown-Lidcombe Hospitals. The core insight that emerged: the maternity and midwifery team at SWSLHD uses a highly specialised lexicon. Standard AI translation tools broke down on this terminology, producing outputs that sounded plausible but were clinically inaccurate.

Phase 2 — Proof of Concept

Advancing to Phase 2 marked a decisive validation: the feasibility had been established, and the NSW Government chose to back the development of a functional proof of concept. This phase moved from research into product: an application featuring real-time, two-way transcription and translation, purpose-built for clinical environments.

A critical development was the construction of custom medical glossaries in partnership with Sam Shen, Health Language Service Manager at SWSLHD. These glossaries encode the precise translations used by SWSLHD's Interpreter Services team for CALD patient populations, ensuring that clinical terms, medication names, procedural language, and culturally sensitive terminology are rendered with the accuracy that clinical safety demands.

The result: a midwife can now conduct an antenatal consultation with an Arabic-speaking patient. The midwife speaks in English. The patient hears Arabic. The patient responds. The midwife reads the English translation instantly, without a third-party interpreter or the risk of family members filtering or softening critical medical information.

What SWSLHD's Experience Means for Your Organisation

If your hospital, health district, or aged care network serves patients from culturally and linguistically diverse backgrounds, the barriers SWSLHD faced are almost certainly familiar. The reliance on ad-hoc family interpretation. The unpredictable interpreter wait times. The discomfort of knowing that clinical nuance is being lost in translation.

VideoTranslatorAI's platform is now available to healthcare organisations that need interpreted telehealth consultations, multilingual in-person support, and structured documentation outputs, all from a platform that has been built to healthcare standards, not adapted from a consumer tool.

Ready to start? Try if for Free

Can this support healthcare and aged care conversations?

Yes. Healthcare and aged care teams can use interpreted sessions to improve communication clarity, then produce structured summaries to support continuity and documentation quality.

How do I choose between In Person, Video Call, and Broadcast?

Use In Person for face-to-face interpreted conversations or spoken note capture, Video Call for English-only transcription or two-language interpretation, and Broadcast when sessions involve 3 or more active languages with one-to-many and two-way participation where needed. If you are uncertain, start with your most common meeting format and language mix, then expand mode usage as your team gains confidence.

Can we trial VideoTranslatorAI before wider rollout?

Yes. Teams can start with the free trial experience to validate workflows, language coverage, and summary output quality in real scenarios. A practical trial should include at least one In Person session, one Video Call scenario, and one workflow using Prompt Customisation so you can confirm fit before broader operational adoption.