Speech-to-Text Interpreting

Speech-to-text interpreters simultaneously translate spoken language into written text. Non-verbal elements as well as everything else that the hearing audience perceives are also transcribed.

It looks something like this:
(Moderator) It is my pleasure to introduce the next speaker, Ms. Jane Doe. 
(Applause)
(Jane Doe) Thank you very much for the invitation! (laughs)

Users can read along in real time. If desired, the transcript can be edited to produce a live transcript

Speech-to-text interpretation is suited to a wide variety of audiences. The text can be displayed for individual, most often deaf or hard-of-hearing users on their screens, enabling them to read along in real time and actively participate in an event. The written interpretation can also be projected on a large screen as subtitles or supertitles for all participants to see. Speech-to-text interpreting facilitates barrier-free communication, especially at educational institutions, during appointments with public administration officials and doctors, as well as at business meetings and conferences.

Speech-to-text interpreters primarily work with speech recognition software (which is known as “respeaking”). In some cases, the conventional method, i.e., typing on a conventional keyboard with the aid of a system of shortcuts, is also used. With both methods, the interpreter adds to the text everything else that the hearing audience perceives, thereby enabling equal, barrier-free participation in the event. As with spoken language interpreting, speech-to-text interpreters always work in pairs, or in threes for longer assignments. One interpreter interprets, while another verifies the output, corrects any speech recognition errors or typos and adds information if needed. After 15 minutes, they take turns, as speech-to-text interpreting requires a very high level of concentration. Users receive a link to a special speech-to-text interpretation platform where they can read along. Speech-to-text interpreters work online, on site and also in hybrid settings (one interpreter is on site, the other online).

In most cases, speech-to-text interpreting is intralingual, i.e., within only one language. Spoken German is interpreted into German text, spoken English into English text. Depending on what is needed, interpreting can be done word-for-word, summarized or simplified from the spoken language into the written language. FF-Translations also works interlingually, whereby spoken German or English is simultaneously interpreted and transcribed into the other language.

Speech-to-text interpretation is used to enable participants to follow an event or lecture in real time. Afterwards, the text is deleted again for data protection reasons. However, if the client wishes to use the transcript after the event, it is possible to create an edited live transcript. Here, any typos and omissions are corrected. The client must first obtain consent from all of the speakers and participants before creating such a transcript.