Real-time voice

Real-time voice enables users to have natural, hands-free voice conversations with Glean Assistant. Instead of typing, users can speak and hear Glean respond out loud, with interruptible back-and-forth dialogue. Conversations are transcribed into standard chats so users can scroll, copy, or continue by text later.

Real-time voice is different from dictation or text-to-speech. It enables low-latency, interactive dialogue where users can interrupt, ask follow-up questions, and receive spoken answers grounded in your organization's knowledge.

note

Real-time voice is available for Glean Universal Model Key customers only. Customer Key support is not yet available.

Availability

Real-time voice is available on:

• Web application
• Desktop application
• Mobile application

Users access real-time voice through the voice button (waveform icon) in the composer.

Configure real-time voice

Admins control real-time voice rollout from Admin Console → Assistant → Real-time voice.

Three rollout options are available:

Option	Description
Off	Real-time voice is turned off for all users.
On only for admins	Only admins can access real-time voice. This is the recommended starting point for evaluation.
On for everyone	All users in your organization can access real-time voice.

Your selection can be changed at any time.

note

The real-time voice admin setting is only visible for Glean Universal Model Key customers.

Document creation from voice

Users can create document artifacts directly from voice conversations. Instead of receiving a transient chat response, Glean generates a persistent document that appears in the Canvas and is saved to the user's Library. This is useful for brainstorming sessions, meeting prep, and other scenarios where spoken conversations produce content worth keeping.

Only document artifacts are supported from voice at this time. Slides, HTML, and other visual artifact types must be created from text chat.

Privacy and data handling

• No raw audio is stored. Audio streams are routed from the user's browser through Glean servers to OpenAI and back. No audio recordings are retained.
• Transcriptions are stored like text chats. Voice transcriptions follow the same storage and retention policies as standard chat messages.
• Permissions are enforced. Glean's existing security model and document-level permissions apply to all voice interactions, the same as text-based chats.

Usage and pricing

Real-time voice usage may be subject to usage-based pricing.

Recommended rollout approach

1. Start with admin-only access: Set the feature to On only for admins to evaluate the experience.
2. Test use cases: Have admins test real-time voice in various scenarios to understand value and behavior.
3. Expand gradually: Once comfortable, enable for all users or specific groups.
4. Monitor feedback: Collect user feedback through standard channels to inform adjustments.

Limitations

• Real-time voice is available for Glean Universal Model Key customers only. Customer Key support is not yet available.
• Only document artifacts can be created from voice at this time. Slides, HTML, and other visual artifact types must be created from text chat.

See also

• Real-time voice user guide - End-user documentation for using real-time voice
• Usage-based pricing - Information about FlexCredits and pricing