Audio Conversation Survey Questions

Use Audio Conversation questions when a study calls for a spoken, AI-moderated interview instead of a typed or single-clip audio response.

When To Use Audio Conversation

Audio Conversation runs a spoken, AI-moderated interview: the respondent talks, the model asks follow-up questions, and its reply is spoken back. It suits qualitative probes, guided reflection, or interview-like prompts where a spoken exchange fits the study better than a typed one — the AI-moderated equivalent of Text Chat, but voice in both directions. It requires a Standard plan or above.

Block Builder: Question Type

Pilot the prompt and every provider before live collection, listen to sample conversations for unexpected AI behavior, and confirm your IRB protocol allows AI-mediated respondent interaction.

How It Supports Research Design

System prompt control defines how the assistant should interact with participants, same as Text Chat.
An optional spoken greeting lets the assistant open the conversation before the respondent's first turn.
Three independent pipeline legs — speech-to-text, the reasoning model, and text-to-speech — can each use a different vendor (ElevenLabs, Gemini, Hume, or an OpenAI-compatible endpoint), either directly or through a saved credential. A provider preset fills all three at once; "Custom" leaves each one manual.
Turn limits and turn duration (minimum/maximum respondent turns, minimum/maximum seconds per recording) keep conversations comparable across respondents.
Playback controls choose whether the spoken reply plays automatically and whether a running text transcript is shown alongside the audio.
A typed-reply alternative can be turned on per question — a low-key "type instead" control a respondent can use for any turn, without justification, if they can't provide a spoken reply. The typed-reply box has its own Character limit, Disable pasting, and Track pasting settings.
Response delay can make the interaction feel paced rather than instantaneous.
Transcript export includes each turn's text and which model produced each assistant reply, alongside the respondent's own recording.

The respondent-turn transcript comes from an automatic speech-to-text step, not from a human — treat it as unverified, the same caution the Audio Response help article gives for its own optional transcription feature. A quiet room, an accent the model handles poorly, or background noise can all produce a transcription error that looks like a "bad answer" in review but is really a transcription one. When it matters, listen to the recording rather than trusting the text alone.

Using a synthetic AI voice with respondents carries disclosure obligations that a typed chatbot does not: several jurisdictions (for example California's AB 2905) require telling a respondent, out loud, that they're speaking with an AI-generated voice, and broader AI-transparency rules (for example the EU AI Act) add requirements around synthetic media generally. Use the system prompt, greeting, or instructions field to disclose this plainly, and check your own institution's and jurisdiction's current requirements before fielding — this article is not legal guidance.

Preview: Respondent Experience

Configure It In Domandata

Step 1: Open Block Builder. Open the survey and select the block where the voice interview belongs.
Step 1: Open Block Builder
Step 2: Add an Audio Conversation question. Choose Audio Conversation from the question type selector.
Step 2: Choose Audio Conversation question type
Step 3: Write the prompts. Add the respondent-facing prompt, the system prompt that controls the interview, an optional spoken greeting, and any on-screen instructions.
Step 3: Write the prompts
Step 4: Configure the pipeline. Pick a provider preset (or Custom), then confirm the speech-to-text, model, and text-to-speech vendors, API keys or saved credentials, and voice.
Step 4: Configure the pipeline
Step 5: Set turns, duration, and accessibility. Set minimum and maximum turns, minimum and maximum recording length, whether replies auto-play, whether the transcript is shown, and whether to allow a typed-reply alternative. Instrument Review warns if the minimum turn length is longer than the maximum.
Step 5: Set turns, duration, and accessibility
Step 6: Preview with realistic responses. Use Preview to test the interaction before publishing, including brief, off-topic, and hard-to-transcribe (noisy, accented, quiet) recordings.
Step 6: Preview with realistic responses
Step 7: Export and review transcripts. Submit a test response, confirm the transcript and recordings are suitable for coding and audit review, and check privacy expectations before live collection — respondent audio is retained as part of the response.
Step 7: Export and review transcripts

Data And Export Notes

Audio Conversation exports the conversation transcript alongside each respondent recording. The assistant's spoken replies are not stored — they're regenerable from the transcript and would otherwise double the size of every response — so only the respondent's own audio is retained. Treat both the recordings and the transcript as qualitative research data, and review privacy, consent, and institutional review requirements before collecting live responses.

A Silicon Sample run leaves this question's column blank rather than fabricating a conversation — Domandata does not simulate a voice interview on an author's behalf, unlike some other question types.

Export: Data and Files

Related Help

Text Chat Survey Questions for a typed AI-moderated conversation instead of a spoken one
Audio Response Survey Questions for a single recorded or uploaded clip instead of a back-and-forth conversation
Content Blocks in Research Surveys
Create Experimental Conditions
Publish and Share a Research Survey