Skip to main content

Voice input

Sometimes you want to ask a question while reading a chart on screen, walking, or just because typing is slower than talking. Qry has built-in voice input: click the mic, speak, the transcribed text appears in the prompt as if you'd typed it.

Goal​

You finish this page with at least one prompt sent via voice.

Prerequisites​

  • A working microphone and the browser's mic permission for your tenant URL.
  • Dictation switched on for your installation. It is on by default; if there is no microphone in the prompt, an admin has switched it off in Admin › System configuration › Platform › Speech-to-text.

Steps​

1. Click the microphone​

The microphone (Dictate) sits in the prompt, next to the main button — on a phone, between the text and Talk or Send. Click it.

The first time, your browser asks for microphone permission. Allow it.

2. Speak, then stop​

While it records, the microphone becomes a solid stop button, and a label above it shows Recording with the time (Recording 0:05) and the level of your voice, so you can see it is hearing you. Speak naturally, then click the stop button (Stop dictation). The label reads Transcribing… for a moment.

You don't need to dictate punctuation: the transcription model adds natural sentence breaks.

Hands-free dictation

On a computer, hovering over the microphone shows a second button, Auto-detect speech. With it on, the microphone listens and starts recording when you speak, and stops on its own after about two seconds of silence. Click the microphone again to stop listening.

3. Review and send​

The transcribed text appears in the prompt. Review for accuracy before sending — voice models have known limitations:

  • Acronyms — MRR may transcribe as MR, empire, or M-R-R. Worth a quick fix before sending.
  • Table or column names — customers_v2 rarely comes through clean. Edit before sending.
  • Names — uncommon names get phonetic spellings.

Press Enter or click Send to submit.

What's happening behind the scenes​

  • The browser records the audio and sends it to Qry when you stop.
  • The backend forwards it to the configured provider: OpenAI Whisper (directly or through your Azure AI Foundry resource) or Google Cloud Speech-to-Text.
  • The transcription is returned as text and inserted into the prompt input.
  • The audio is not kept — only the transcribed text becomes part of the conversation.

Common issues​

There is no microphone in the prompt. Dictation is switched off for your installation. Ask an admin to switch it on in Admin › System configuration › Platform › Speech-to-text.

Browser keeps asking for mic permission. Your browser is set to ask every session. Open the tenant URL's site permissions and grant mic permanently.

Transcription cuts me off mid-sentence. With Auto-detect speech on, a pause of about two seconds ends the recording. Speak more continuously, or turn auto-detect off and stop the recording yourself.

Wrong language detected. Short clips are hard to auto-detect. Ask an admin to set a default language for dictation.

Background noise garbles the transcription. The provider does its best with what it gets. Use a quieter environment or a headset mic.

See also​