Skip to content
vDynamiq Docs

Languages & voice

vDynamiq studies are narrated and can be multi-language. This page explains how language and voice work across a study, and how vDynamiq keeps the underlying measurement consistent no matter which language a respondent hears.

Languages are a study-level setting

Section titled “Languages are a study-level setting”

The set of languages a study supports is defined at the study level. When a study has more than one language, vDynamiq produces the respondent-facing content — including scene narration — in each of them. Respondents experience the study in a supported language, and the scenes they see are narrated accordingly.

Voice narration

Section titled “Voice narration”

Each scene has a narration — a spoken voice-over that carries the story. When you build or edit a scene, narration is generated for every language the study supports, so switching a study’s language doesn’t leave gaps. Narration is one of the five parts of a scene you can review and regenerate in the scene editor; see Generating assets for how it’s produced.

The Survey Buddy companion, when enabled, also speaks in keeping with these settings, so the whole experience feels cohesive.

Voice answers from respondents

Section titled “Voice answers from respondents”

Some interaction types let respondents answer by speaking rather than tapping — for example voice_free (spontaneous recall) and tap_speak (elaborating on a choice). On mobile, voice answers can be captured as the respondent speaks. See Voice & streaming for how this behaves during a live survey.

Measurement stays consistent across languages

Section titled “Measurement stays consistent across languages”

A key guarantee: language changes the words a respondent hears, not what is measured. A metric is defined once at the construct level and measured on the same scale regardless of language. Internally, spoken and translated answers are normalised so that analysis treats them consistently — so a multi-language study still rolls up to a single, comparable set of construct scores in aggregation.

Practical notes

Section titled “Practical notes”
  • Add languages at the study level, then review that each scene’s narration is present for each language.
  • Regenerate narration for a scene if you change its wording, so the audio matches the latest text.
  • Voice answers depend on the interaction type and are captured during the survey; the measurement they feed is language-independent.

Where to go next

Section titled “Where to go next”