An audio object stores a recording in your space. You can play it back, add notes, tag it, and embed it like any other object.

Capacities Believer users with AI enabled can run Audio analysis on any audio object for a transcript and a summary.

Audio properties

Every audio object comes with properties that help you describe, organize, and query your recordings:

• Title – The name of the audio object.
• Tags – Labels you can assign to group related recordings.
• Blocks – Where you write your notes on the recording.
• Category – A predefined classification used for filtering and queries.

You can set these properties yourself. Believers with AI enabled can use Audio analysis to add a transcript and a summary. The transcript and summary are searchable.

Supported file types

You can upload audio files with these extensions: *.mp3, *.ogg, *.oga, *.wav, *.m4a, *.webm, *.aac.

Like other media, each uploaded file must be under 100 MB.

Record or add audio

Record in Capacities

Recording is available in early access for Capacities Believer. After early access, it will be part of Capacities Pro.

You can start a recording in several ways:

• On desktop: New > Audio, or the microphone icon in the daily note
• On mobile: the Record audio action in the app
• From your phone: the Record audio widget (iOS), or the matching quick action when it is available

Each recording is saved as an audio object in your space.

Recording length and format

There is no fixed time limit. Recordings are capped by file size (upload limit 100 MB).

As a rule of thumb you should get at least about 90 minutes, and often more, depending on the device and how dense the audio is. Capacities warns you when the recording is getting close to the size limit, then stops safely so the recording is preserved.

On macOS desktop you can optionally include system audio with the microphone. Windows and mobile record the microphone only.

Send a voice note from WhatsApp or Telegram

You can also create an audio object by sending yourself a voice note with the WhatsApp or Telegram integrations. The voice note is added to your daily note and saved as an audio object.

Playback

On an audio object you can:

• Change playback speed
• Repeat
• Change volume
• Download the file

Changing the title

If you want to change the title of your audio object so it is easier to link to elsewhere, delete the title and replace it.

Adding notes to your audio file

Each audio object has a notes section. Use it for key points, or to link the recording to other objects.

Audio analysis

Audio analysis is part of Capacities' media analysis feature. It helps Capacities understand the content of a recording so you can search it and use it as context in AI chat.

Audio analysis is available in early access for Capacities Believer. You also need AI features enabled in the space. After early access, it will be part of Capacities Pro.

You can analyze any audio object: recordings you make in Capacities, voice notes from WhatsApp or Telegram, and other supported audio files in your space.

Open any audio object and click Analyze. Capacities processes the recording and fills in:

• Transcript – The spoken words as text.
• Summary – A short overview of the recording.

The transcript and summary are searchable. They are also available as context for the AI assistant.

If you want to run analysis again, use Analyze on the audio object. Progress and errors are shown in place.

Analysis uses your AI budget. Longer recordings use more budget. Check remaining budget in Settings > AI Assistant.

If your AI budget is exhausted, analysis shows an AI budget used state and opens AI settings. Wait until your budget resets at the start of the next billing month, or add Pro+ or Believer+ for a much larger monthly budget and top-ups when you need more. See Subscriptions and payments for details.

FAQs

• How do I add or record audio?
  On desktop, use New > Audio or the microphone in the daily note. On mobile, use Record audio in the app or the Record audio widget. You can also send a voice note from WhatsApp or Telegram.
• How long can I record?
  There is no exact minute limit — recordings are limited by file size (100 MB). You should get at least about 90 minutes, and often more. Capacities warns you when the size is approaching the limit. See Recording length and format.
• Can I analyze a WhatsApp or Telegram voice note?
  Yes. Any audio object can be analyzed when AI is turned on for the space (Believer early access). Open the audio object and click Analyze.
• Do I need AI to record audio?
  No. Recording creates an audio object without AI. Audio analysis needs AI turned on.
• Do I need to analyze audio to play it back or add notes?
  No. Playback and notes work on every audio object. Analysis adds the transcript and summary.