Automated transcription is fast becoming a staple of production processes for media organizations. The tedious manual process of listening back to recordings has been replaced by fast, efficient workflows that enable teams to quickly find key moments in their audio and video and create new content. But while AI-powered tools are competing in this transcription space, few offer help dealing with events in which multiple languages can be spoken. Solving this challenge has been identified as one of the most desired innovations in the future of voice recognition.
For news organizations to cover global events such as political summits, conflicts or sporting events in multiple languages, they are turning to manual transcribers and translators, which can be a time-consuming and expensive process. The same is true for businesses of all kinds across the world who work and share information across borders. Trint’s innovation is to remove the language barrier that gets in the way of business and media communication –making workflows seamless in one, two or a dozen languages.
At Trint, we understand the struggles that news organizations face as our founder, Jeff Kofman, experienced it himself during a 30+ year career as a journalist and war correspondent with ABC, CBS and CBC News. It’s this insight that helped us develop our live transcription feature that allows journalists to capture moments in over 40 languages from their phone or broadcast stream. With additional AI translation into another language available after the transcription is complete.
Trint’s latest innovation, Multi-Language Transcription and Translation Live (MLTT Live), is now set to truly disrupt this space. It uses AI to automatically recognise and transcribe the languages being spoken in live audio/video streams, accurately transcribe them and then translate the text into the journalist's chosen language.
Each word in the transcript will also be linked to the source audio/video, allowing journalists to playback and listen to what was said, verify the accuracy of the transcript and amend any text in the moment. The result: news organizations can deliver live reports of the biggest global events with confidence that their quotes are 100% accurate.
Once a journalist has their live transcript, MLTT Live integrates with the rest of Trint’s collaboration features to streamline the content workflow. Transcripts and translations are searchable and editable, allowing teams anywhere in the world to simultaneously review, correct and publish quotes – even while the transcription is still live.
Trint’s current live transcription feature has already revolutionized live news coverage by saving media organizations 75% of time taken when reporting on events with one language. MLTT Live will extend this to events where multiple languages are spoken and where the journalist(s) cannot speak all of those languages. Creating a more efficient workflow that saves media organizations both time and money.
Institutions like the EU (with 24 official languages) and the UN (with 6) have a commitment to providing simultaneous interpretation for attendees, including journalists, but this is expensive and smaller organizations cannot make the same commitment. With Trint’s innovation, any journalist at any event will get simultaneous interpretation into any language.
“It's fantastic to get such an endorsement for our plans from the UK Government through Innovate UK. The grant will enable us to accelerate important innovations to help us to continue to build AI products that are truly distinct, world-beating, and valuable to our customers.”
Jamie Morrison, Head of Product, Trint
