| Term ID | Term Name | Definition | Broader | Release Date |
|---|---|---|---|---|
| 1 | OCR | OCR is intended to represent text that has been captured by an Optical Character Recognition process - such as a scanned script, or from existing bitmap subtitle representations e.g. DVD, DVB or ‘burnt in’ on screen text. | ||
| Term ID | Term Name | Definition | Broader | Release Date |
| 2 | ASR | ASR is intended to represent text that has been captured by an Automatic Speech Recognition (ASR) process based on audio from the programme content. The ASR could be performed by a specific recognition engine capable of taking an audio signal and extracting any spoken utterances contained therein and translating them to an accurate textual representation. | ||
| Term ID | Term Name | Definition | Broader | Release Date |
| 3 | Steno | Steno is intended to represent text that has been captured via a Stenography process - using trained stenographers, bespoke keyboards and keystroke translation software. | ||
| Term ID | Term Name | Definition | Broader | Release Date |
| 4 | Typing | Typing is intended to represent text that has been captured by typists using standard Qwerty or similar keyboards; it is possible that certain Typing scenarios may use multiple keyboards and typists to match the speed of the spoken word in the audio signal. | ||
| Term ID | Term Name | Definition | Broader | Release Date |
| 5 | Respeaking | Respeaking is intended to represent text that has been originated by trained respeakers dictating into a speech recognition engine trained to their own voice. Respeakers repeat the spoken words in the audio signal almost synchronously with the content and add spoken punctuation and other presentation commands as required. | ||
| Term ID | Term Name | Definition | Broader | Release Date |
| 6 | Import Script | Import Script is intended to represent text that has been originated in external systems and imported as a text-based file for use in a subtitling authoring system. | ||
| Term ID | Term Name | Definition | Broader | Release Date |
| 7 | Extract | Extract is intended to represent subtitles that have been extracted from files or sources that also include audio or video; for example data embedded within wrapper-based formats, i.e. MXF containing SMPTE 436M subtitles, ‘tall’ video with teletext, or other related formats. | ||
| Term ID | Term Name | Definition | Broader | Release Date |
| 8 | Transcoded | Transcoded is intended to represent subtitles that have been taken from an existing subtitle only data in an alternative format (which could be either a file or stream) and modified or transcoded to an EBU-TT document. Transcoding may be a lossy process, and may result in substantive, irreversible changes to the subtitle data. |