NOTE: New users looking for instructions for how to get started recording with ModelTalker should begin at this link.

Important Note: Be sure to fully quit Chrome before following the instructions below if you plan to use the web recorder. It is best to set your audio controls first, then start Chrome.

These instructions are based on macOS 10.14; the procedure for other versions may differ in some specifics, but should be similar. These checks should be done at the start of EVERY recording session–you may want to print out this checklist so that you can easily refer to it every time. Having consistent audio settings is crucial to successful voice banking and your audio settings can get changed without you knowing it. Other programs that use audio, and even System updates can possibly result in settings changes.

Get your computer ready

You want a “clean slate” computer each time you go to record for voice banking. Make sure there are no programs except Chrome running on your Mac. Not even Mail or Preview, and especially not iTunes or FaceTime. Moreover, you should not have any extra tabs open in Chrome. For instance, if you use Chrome to read Gmail, close the Gmail tab while recording.

Do not set your Mac to run system updates automatically, and if you are using a virus scan program, make sure it is not downloading updates or doing a full system scan while recording.

Basic audio settings

  1. Plug in your USB headset mic.
  2. Open System Preferences.
  3. Click on “Sound” (the speaker icon).
  4. Click on “Input.”
  5. Select your headset microphone as the Input device.
  6. Slide the input volume control all the way to the right. (You may need to reduce it if you consistently get warnings that your speech is too loud and microphone adjustments do not help. If so, make sure that it is at the same level for all recording sessions.)
  7. Try speaking at a normal volume with your headset on and check that the Input Level indicator moves when you speak.
  8. Click on “Output.”
  9. Make sure your headset is selected as the output device.
  10. Adjust the output volume to about mid-scale (you will be able to adjust the output level with the keyboard controls if this setting is not ideal when you start to record).
  11. Exit from System Preferences.

Microphone positioning

Another way to improve audio consistency is by making sure you have positioned the microphone exactly the same way for each recording session. We strongly recommend using a headset with attached microphone as one way to help ensure consistent microphone positioning. However, even with a headset microphone, it is necessary to take care that the microphone is positioned exactly the same way each time you record.

Never position the microphone directly in front of your mouth. Correct microphone placement should have the microphone parallel to your lips, close to your face and far enough to the side of your mouth to ensure that direct airflow from your mouth and nose cannot hit the microphone. Most headset microphones have flexible booms that can be bent to assist in correct positioning. Although the microphone should be as close as possible to your face, don’t allow hair, whiskers, or anything else to brush against the microphone while recording.

Speaking style and rate

Finally, you need to be as consistent as possible in your speaking style and speaking rate from one recording session to the next. To achieve this, we recommend starting each new session by first listening to a sample of recordings from your previous session to hear the way you sounded during that session. Then record just a few new sentences and listen to them and compare them with some previous recordings. Ask yourself:

  • Does my voice sound the same?
  • Is my speaking rate the same as before?
  • Is the loudness the same?
  • Is the overall audio quality the same as before?

If the answer to any of these questions is “No,” try to adjust to reduce the differences and if you cannot fix the issue don’t record. Maybe letting your voice rest, or taking a drink of water will help if it is an issue with your speech quality or speaking rate. If there are problems you can’t fix, let us know so we can try to help.