Podcast Checker

Audio Audit’s Podcast Checker is an automated quality checking tool for podcasts. Whether you’re a DIY podcaster, work with a production company or run a recording studio, it checks every episode for loudness, noise, silences, swearing, encoding and metadata, and shows you how to fix whatever it finds. If you record, edit and produce your own show, it’s a second pair of ears that doesn’t get tired.

If you work with production professionals, the Podcast Checker gives you a number for each episode instead of a leap of faith. Even skilled producers overlook a standard or a missing tag now and then. Audio Audit measures the listener experience and the metadata, so you can see what was delivered. Most podcasts we check turn up at least one warning.

If you’re part of a recording studio or production company, your reputation depends on consistent output. The Podcast Checker reviews each episode against a standard, and you can create custom standards with your own limits. Every report gets a Passed or Failed verdict, so you can show a client that an episode meets spec.

You can upload a file or use any public RSS feed to get started, and everything is managed in your web browser. Each report links to guides on the checks that need attention, so you can fix problems and help your show sound right on Apple Podcasts, Spotify and everywhere else it’s published.

Signing up for the free plan takes a minute, needs no credit card and gives you enough credits to run several reports each month. If you turn on automatic checking for your podcast, you’ll receive a report by email whenever you publish a new episode, and you’ll be told if your feed goes down.

Checks that are run

These are the checks that are currently being run whilst generating each report. We have detailed resources for most topics if youʼd like to learn more. If you start by generating a report then you can ignore the topics where youʼre already performing well.

Metadata

• Metadata tags — embedded information such as title, album name, artist name and track number
• Cover image — artwork shown with an episode
• Chapters — easier navigation to segments within an episode

Signal encoding

• Bit rate — data to store each second of audio
• Number of channels — stereo or mono
• Sample rate — how many times per second a waveform is sampled
• Sample width — number of digital steps available to represent the audio signal

Audio quality

• Loudness (LUFS) — industry standard for measuring perceived loudness to avoid sudden jumps in volume between podcasts
• Peak volume — maximum signal strength within a file which indicates possible clipping
• Noise floor — a high value indicates poor quality microphones or forgotten noise reduction filtering
• Quiet channel — is the left or right channel missing in a stereo file?

Editing mistakes

• Silences at beginning and end — thereʼs a recommended time gap to leave before and after
• Unwanted noises — maybe you or a guest coughed, a dog barked or a phone received a notification
• Silences in the middle — sometimes this can happen if something got cut out or a clip/jingle didnʼt make it to the final export
• Restarted sentences — uses speech recognition to find similar repeated sections that were likely forgotten about
• Swearing detection — uses speech recognition to protect shows that are published as suitable for younger listeners

Feed

• Valid RSS — validity of feed is checked while automatic reporting is enabled so youʼll know if something breaks on you or your host's website and syndication systems
• Slow server response — the response time is logged each time it is checked and you can see this data as a graph and as averages over the past day, week and month, ensuring high availability
• Episode download speed — soon there will also be a log of download times and speeds of each episode so you compare this with others and keep load at a reasonable level