Advanced Voice Mode on ChatGPT features more natural, real-time conversations that pick up on and respond with emotion and non-verbal cues.

Advanced Voice Mode on ChatGPT is currently in a limited alpha. Please note that it may make mistakes, and access and rate limits are subject to change.

If you are invited to the alpha, you will receive an email with instructions on how to use Advanced Voice mode. You will also see a tooltip on the bottom-right inviting you to try advanced Voice Mode when you open the app.

To start a conversation with advanced Voice Mode, select the Voice icon on the bottom-right of the screen:

Once you begin a conversation with advanced Voice Mode, you will be taken to the following screen:

You can mute or unmute your microphone by selecting the microphone icon on the bottom-left of the screen. You can end the conversation by pressing the red icon on the bottom-right of the screen.

At any time, you can switch to the standard Voice Mode by selecting Advanced at the top of the screen.

Please note that you will need to provide the ChatGPT app Microphone permission to use this feature.

Usage of advanced Voice (audio inputs and outputs) is limited on a daily basis, and precise limits are subject to change. The ChatGPT app will output a warning when you have 3 minutes left of audio.

Once the limit is reached, the conversation will immediately end and you will be invited to use our standard voice mode.

No, advanced Voice Mode currently cannot create memories or access previous memories yet. Advanced Voice Mode also does not have access to custom instructions.

Advanced Voice Mode conversations can be resumed in advanced Voice, text, or standard Voice. Because advanced Voice does not yet support capabilities like memory and custom instructions, conversation with text or standard Voice conversations cannot be resumed in advanced Voice Mode.

Occasionally, interruptions may happen during a conversation with advanced Voice Mode. We recommend using advanced Voice Mode with headphones.

On iPhone, enabling Voice Isolation mic mode can help to avoid unintentional interruptions. You can enable Voice Isolation by opening your Control Panel while using advanced Voice, selecting Mic Mode, and switching to Voice Isolation.

If you are still experiencing issues, we recommend closing your app and restarting, turning up the volume of your assistant, or moving to a quieter environment.

Please note, the advanced Voice Mode experience is not yet optimized for use with in-car bluetooth or speakerphone.

Advanced Voice Mode is not yet available for use with GPTs.

No. To respect creators’ rights, we’ve put in place several mitigations, including new filters, to prevent advanced Voice Mode from responding with musical content including singing.

Video and screensharing support is not yet part of the advanced voice mode alpha, but will be available at a later date.

We are planning for all Plus users to have access in the fall. Exact timelines depend on meeting our high safety and reliability bar. We are also working on rolling out the new video and screen sharing capabilities we demoed separately, and will keep you posted on that timeline.

Yes, advanced Voice Mode is only available to a limited number of users on Plus accounts.

Advanced voice conversations with GPT-4o are inherently multimodal, allowing for audio exchange between you and the model. As a result, when this audio is transcribed, the transcription might not always align perfectly with the original conversation.

During the alpha, we will use audio from conversations with advanced Voice Mode to train our models if you have shared your audio with us. You can opt out of audio training by disabling “Improve voice for everyone” in your Data Controls Settings.

If you do not see “Improve voice for everyone” in settings, then you haven’t shared your audio with us and we will not use your audio for training our models.

With Standard Voice Mode, If you choose to share your audio, then we will store audio from your voice chats rather than deleting audio clips once transcription is complete. We will take steps to reduce the amount of personal information in audio from voice chat that is used to train our models. Our team may review the audio that you’ve shared with us.

Learn more about sharing your audio to improve voice chats in Standard Voice Mode.