Voice and dictation
Understand speech input, speech output, and the services behind them.
On this page
Dictation and spoken replies are separate
Speech-to-text turns microphone audio into text, which is dictation. Text-to-speech turns a reply into audio. Configuring one does not configure the other, and a build may expose dictation without a complete spoken conversation workflow.
Check which voice features your installed client exposes before you spend time on a service credential.
you know which of the two you are setting up, and only configured that one.
Check speech input
-
Configure a speech-to-text provider
Do this on the host, in the component that owns the setting.
-
Allow the microphone
The client asks for microphone access. Say yes on the device you are dictating on.
-
Record something short and harmless
One ordinary sentence. Nothing private.
-
Read the transcript before sending it
Check the words came out right before they go to your coding assistant.
If recording works but transcription fails, check the speech provider’s credentials, endpoint, model and quota. If no audio is captured at all, check microphone permission and the selected input device first.
your sentence appears as correct text in the message box.
Voice input allowance
Managed voice input has a plan allowance, counted in minutes of audio you dictate.
| Plan | Per account, each UTC day |
|---|---|
| Free | 5 minutes |
| Lyre Pro | 15 minutes |
| Lyre AI | 60 minutes |
The allowance resets at 00:00 UTC. On every plan a single recording is limited to two minutes, and technical size, rate and concurrency limits also apply.
The client shows the remaining allowance and disables capture until the daily reset.
you can see the remaining minutes in the client before you start a long dictation.
Audio may leave the device
A hosted transcription service receives your audio in order to process it. Choose the provider with that in mind, avoid dictating secrets, and correct names or commands before sending a transcript.
See integrations for the distinction between coding, transcription and other connected services.
you know which company receives the recording before you press record.