Roadmap
The project is under way with priorities influenced by EBU members and other contributors. The following are some of the possible areas of work for the future:
Additional vendors
Adding support for additional vendors: cloud, on-premise, self-trained models. If you currently integrate with a vendor that has particular expertise with a language not served by the larger cloud vendors, we would be particularly interested.
Tool support for multiple speakers
Ability to identify and amend speaker detection in the EuroVOX Tool. Add additional metadata for the speaker, along with possible links to external personality stores.
Additional timeline metadata
A mode in the tool to be able to visualise and edit additional metadata associated with the content timeline. Beyond transcription and translation with multiple speakers: time markers, regions, segments, topic markers, etc. Having a single tool to be able to work with content metadata.
Collaborative working
Sharing content between users, teams and organizations. Enabling multiple users to work on the same job with visible changes - different teams editing different translation languages.
Advanced audio handling
Being able to more effectively deal with mixed-language content, e.g. by automatic detection from a set of candidate languages, or by marking up regions as being from a different language.
Better handling of multichannel audio: different audio tracks for different languages, audio tracks for different speakers, same speaker on different microphones. The ability to manage how content is processed beyond simple mono or stereo.
Live Mode and user-controlled outputs
The ability to set up a live transcription, view it happening, make corrections and then output to an external service as a way of enabling realtime transcriptions in your user workflows.