Conserving your audio generation quota

Audio.co relies on a third-party service to turn your script into a broadcast-ready bulletin. This is why each plan includes a usage quota outlining how many seconds you and your team have remaining.

Quotas reset to zero at the end of your current billing period, so it will be a month after you first signed up for Audio.co rather than calendar months.

There are ways you can conserve your team's available seconds of audio and make them go further:

  1. Don't start generating editions on a schedule until you're ready to broadcast them. Once a schedule is set up, editions will automatically be created as instructed. If you're still setting things up, it's best to generate editions manually.
  2. Use your own news sources. If you set up an edition to pull content from a feed you don't manage, there can be a high volume of new stories, meaning by the next time an edition is generated, Audio.co won't be able to re-use any previously generated audio to help conserve your available seconds.
  3. In the news sources you are using, make sure the 'publication limit' is disabled. When this feature is enabled after a set time period has passed, Audio.co will rewrite a story's script to refresh it, generating new audio each time. Usually, Audio.co generates new audio only when the script changes; otherwise, it reuses the previously generated audio to conserve your available seconds.
  4. On your edition's scheduling page, ensure you do not generate a new edition when a source updates. This option is fine if you're only using RSS feeds you curate yourself, but otherwise, if you're using an external source, any updates to their feed could generate a new edition - even if it's just updating an existing story. This means that rather than generating at the times you want, it could produce a dozen editions that may not even be used.
  5. Create your own AI model that is a bit snappier than the default. If your AI model is trying to write stories two or three sentences long instead of four, you'll generate fewer seconds of audio each time. (Make sure you add plenty of good examples so the AI knows what to work from)