How to Create Same-Language Captions for Deaf and Hard-of-Hearing Theatre Audiences

What same-language captions are—and are not

Same-language captions give Deaf and hard-of-hearing theatre audiences a text version of the performance in the language being performed on stage. They normally display dialogue, speaker changes where needed, and relevant non-speech information such as sound effects and off-stage noises. The Edinburgh Festival Fringe describes captioning as dialogue, sound effects and off-stage noises in a text format, normally created by an operator, and notes that it can also benefit people with learning disabilities and people for whom English is not a first language (Edinburgh Festival Fringe). Stagetext frames the aim more broadly: subtitles should be a visual soundtrack to the spoken word, giving equal access rather than a summary (Stagetext, Digital Subtitling Guidelines).

They are not the same as surtitles. The Fringe distinguishes captioning, which displays dialogue and sounds, from surtitles, which are for productions performed in a foreign language and include the English translation, usually edited, and may not include sounds that are not speech (Edinburgh Festival Fringe). In other words, a surtitle is usually a translation; a same-language caption is an access text for the language already being performed. The distinction matters because captioners need different editorial priorities: surtitles often compress for reading speed across a language barrier, while same-language captions should preserve the spoken style and access to sound.

Start with the script, not the screen

Good captioning begins before the technical rehearsal. Stagetext advises working from a copy of the text where possible, or from a dialogue list and a glossary of atypical words and special references. The preparation should identify script, speaker and place names, specialist vocabulary, and sound effects (Stagetext, Digital Subtitling Guidelines). Ask the director or production team for the version that will actually be performed: cuts, reordered scenes, ad-libs, pronunciation notes, and recorded sound cues all affect the caption file.

The default should be verbatim, styled as the person speaks, and, where possible, each caption syntactically self-contained (Stagetext, Digital Subtitling Guidelines). Verbatim does not mean careless line breaks. A caption that is literally accurate but split across an awkward clause can be harder to read than a slightly reflowed line. Stagetext recommends an average subtitle length of 42 characters, no more than two lines, and, where possible, a longer bottom line so the caption interferes less with the image (Stagetext, Digital Subtitling Guidelines). Those figures are working norms, not an excuse to cut meaning.

Verbatim, timing and minimal editing

Stagetext’s guidelines set verbatim styling as the default, but they also recognise that timing may require editing. When dialogue must fit a set time, the text must remain coherent, and unnecessary words such as really, just, perhaps, that, quite, who, was, or which may be removed (Stagetext, Digital Subtitling Guidelines). The practical rule is: edit only when the display or reading speed requires it, preserve meaning and character voice, and record the decision. Do not silently standardise dialect, remove repetition that is dramatically meaningful, or turn overlapping speech into a tidy single stream unless the production team has agreed that solution.

What to include beyond dialogue

Non-speech information needs its own editorial policy. The Fringe guidance includes sound effects and off-stage noises as caption content (Edinburgh Festival Fringe). Stagetext advises that descriptions should be fact based and remain in the world of the play or video (Stagetext, Digital Subtitling Guidelines). In practice, this means captioning a door slam, a phone ringing, a recorded announcement, or a significant musical cue when it carries meaning, but avoiding interpretation such as ominous music unless the production brief explicitly asks for that interpretive layer. For Deaf and hard-of-hearing audiences, missing sound cues can remove plot information, comic timing, or spatial orientation.

A production workflow

The following workflow assumes a scripted production with a prepared caption file and an operator, though the same principles apply to simpler setups. It is a workable division of labour rather than a universal standard.

1. Agree the access brief. Clarify who the captions are for, which performances will be captioned, whether the display will be a shared caption unit or personal tablets and phones, and who approves the final text. Stagetext’s resources describe captions displayed on a caption unit, tablet or phone, or online (Stagetext resources).
2. Collect the performance text. Obtain the script, dialogue list, glossary, song lyrics if any, and a sound cue list. Mark speaker names, unusual vocabulary, foreign-language phrases, and non-speech sounds. Stagetext’s guidance specifically lists script, speaker and place names, specialist vocabulary, and sound effects as reference points (Stagetext, Digital Subtitling Guidelines).
3. Draft the caption script. Work line by line, keeping captions short and self-contained. Use the 42-character average and two-line maximum as a formatting discipline (Stagetext, Digital Subtitling Guidelines). Label speakers only when the audience needs the information; constant labels can clutter a fast scene.
4. Cue the file. Create cue points against the stage manager’s book or a rehearsal recording. Note the exact trigger—an actor’s line, a sound cue, a lighting change—and the expected operator action. A cue sheet is not a duplicate script; it is the operator’s map through the performance.
5. Rehearse the display. Test sightlines, brightness, latency, and legibility from the audience positions where captions will be used. If personal devices are used, test charging, network reliability if relevant, and how audience members collect and return them. A caption that is accurate but unreadable is not access.
6. Run the mute check. Stagetext recommends watching the whole video on mute and asking whether the captions provide equal access (Stagetext, Digital Subtitling Guidelines). For live theatre, the equivalent is to review a run with sound reduced or removed and ask whether dialogue, laughter, music cues, and off-stage noises remain intelligible through the text.
7. Review and archive. Keep the approved caption script, cue sheet, sound list, and operator notes together. Version control matters for remounts, tours, and co-productions, when a production may be revived with a different cast or venue.

Roles and deliverables

On a small production, one person may combine captioner and operator. On a larger one, the work is clearer when separated. The captioner prepares and edits the text; the operator follows the performance and triggers cues; the stage manager supplies accurate timings and changes; the director or dramaturg resolves textual questions and approves cuts; and an access coordinator liaises with audiences and venue staff. The exact titles vary, but the responsibilities should be named in the production schedule rather than left to the technical rehearsal.

Useful deliverables include a clean caption script, a cue-by-cue operator sheet, a sound and non-speech list, a pronunciation guide for names and specialist terms, a display setup note, and an archive copy. These are practical documents, not bureaucracy: they make captions repeatable when a show travels or is remounted, and they allow a new operator to take over without rebuilding the file from memory.

For improvised, participatory, or rapidly changing material, a pre-prepared file may be insufficient. A live captioner or live speech-to-text operator can follow the performance, but latency and error correction become part of the access design. Test the system in the actual room, agree a protocol for corrections, and tell the audience how the captions will work. The same quality questions apply: can readers keep up, and do they receive the dialogue and relevant sounds rather than a delayed summary?

Quality problems to watch for

There is limited publicly available theatre-specific captioning guidance, and much practice adapts principles from television, video, web, and educational captions (University of Melbourne Student Union). That makes editorial judgement important. Common problems include captions that lag behind the voice, lines that break in the middle of a name or phrase, missed sound cues, over-editing that flattens character voice, and displays placed where audiences cannot comfortably read them. A caption file that works on a desktop monitor may fail in a dark auditorium, on a small phone screen, or from the back of a raked seating bank.

The strongest quality check is a structured review: read the caption script aloud at performance speed, watch a run with the sound muted, and ask a small group of Deaf and hard-of-hearing readers to comment on clarity and timing where resources allow. Accessibility is not only a technical output; it is a relationship with the audience. Captioners should be willing to adjust phrasing, display position, or cue timing when the access experience requires it.

Same-language captions across borders

For international work, the phrase same-language is a useful discipline. A production in French needs French captions; a production in British Sign Language and spoken English may require a decision about which languages and sound information are represented, and how those changes are labelled. The caption file should travel with the production as an access document, alongside the script and cue sheet, and should be checked against the version actually performed. If a show is remounted in another country, do not assume the original display hardware, venue sightlines, or operator workflow will transfer unchanged. Verify the technical setup with the receiving venue and retain the language of the performance rather than substituting a translation unless a translated caption track is deliberately commissioned.

Language labelling matters in multilingual productions. If a scene moves between languages, the caption track should make that movement clear without assuming that every audience member reads every language. A same-language file may therefore need short language labels or a consistent typographic convention. These choices should be made with the creative team and, where possible, with audience consultation, because they affect how the performance is understood rather than merely displayed.

A practical checklist

• Identify the performance language and the intended caption users.
• Obtain the performed script, dialogue list, glossary, and sound cues.
• Mark speaker changes, specialist vocabulary, foreign-language phrases, and non-speech sounds.
• Keep captions to an average of about 42 characters and no more than two lines where possible (Stagetext, Digital Subtitling Guidelines).
• Default to verbatim, styled speech; edit only for coherence and reading speed, and record changes.
• Build a cue sheet and rehearse with the operator and stage management.
• Test display position, brightness, latency, and sightlines from audience seating.
• Run a mute check and revise the file before the captioned performance.
• Archive the caption script, cue sheet, sound list, and setup notes for remounts and tours.

Same-language captions are sometimes treated as a late technical add-on. They are better understood as a dramaturgical and access practice: a second text that runs alongside the performance, shaped by the script, the sound world, the venue, and the audience. Done well, they do not compete with the stage. They make the stage available.