Voice Over Audio Testing Techniques: Ensuring High-Quality Recordings

In the realm of audio and video production, the importance of high-quality audio cannot be overstated. Great audio isn’t just important. Most people who watch videos say they are more likely to stop watching a video with bad audio vs. This isn’t that surprising when you think about it. Muddy, muffled, or otherwise garbled or difficult-to-understand audio tracks are frustrating to viewers. Achieving a professional-sounding voice over requires careful attention to various aspects, from recording techniques to post-production editing. This article delves into essential voice over audio testing techniques, including the use of Perceptual Objective Listening Quality Analysis (POLQA), audio gain adjustment, and Voice User Interface (VUI) testing.

Understanding POLQA: Perceptual Objective Listening Quality Analysis

Perceptual Objective Listening Quality Analysis (POLQA) is the latest (2011) objective measurement for speech quality standardized (ITU-T P.863). POLQA is an enhancement to the PESQ algorithm (ITU-TP.862) introduced in 2003 which allowed perceptual speech quality measurements to become a mainstream method to evaluate end-to-end speech clarity. Perceptual speech quality measurements are needed when speech is compressed such as in a voice codec (e.g., G.729, G.711, G.722, AMR-WB, EVS, etc.) and then transported over wireless or IP-based networks.

The principle is similar to PESQ - comparing the clean reference file with the degraded file from the network. But the test methodology is different. The POLQA algorithm was introduced to deal with some of the weaknesses inherent in PESQ, as well as to address wideband telephony or HD voice (i.e., voice communication above 3.4kHz). POLQA has two models, one for narrowband and one for super-wideband. The POLQAv3 goes to full-wideband, the full range of the human ear, and can be used for any network from 300-24,000Hz analog bandwidth.

Why Is POLQA Useful?

The main driver for the new POLQA standard was wideband telephony, sometimes called HD-voice as defined by the G.722 and G.722.1 (wideband AMR). In addition, POLQA analyzes speech to the limit of the human ear (24kHz analog bandwidth).

Essential Voice Over Recording Techniques

So how do you record voice overs that grab and keep your audience’s attention? Here are some techniques:

1. Preparation and Scripting

Not all videos need a ton of preparation. Quick one-off screencasts or a fast demonstration of a new user interface for a colleague can probably be done on the fly. Writing a script is probably the single most important thing you can do to ensure your voice over sounds professional. The best scripts will include everything you intend to say, word for word. And if you don’t want to write a script, you don’t have to! Audiate’s Generate Script feature lets you enter a prompt and writes the script out for you. You simply need to edit it to your liking and you’re done.

2. Creating the Right Environment

You want a space that’s free from distracting noises and where you are unlikely to be interrupted. For example, you might find the quietest place you can find to record is in your car (with the engine off, of course). Next, you need a decent microphone. What we will say, however, is you should definitely avoid recording your voice overs on your laptop microphone, if possible. You don’t need to spend hundreds of dollars, either. You can get a decent USB microphone for $50-$100, which is well worth the investment if you’ll be doing a lot of voice over work. If you do intend to use an external mic, we’d also recommend investing in a pop filter.

3. Microphone Placement and Audio Levels

You also want to consider where to place your microphone. If it’s too close to the person speaking, it will pick up all kinds of weird mouth noises and air puffs. Ideally, place the microphone about six to eight inches from your mouth, and slightly below your chin. You should also be mindful of the surface where you place your mic. Getting the volume level right for your audio ensures that it’s easily heard and not distorted. If the volume is too low, your audience will have trouble hearing what you say. In audio editing, 0db is actually the maximum you want to achieve. Usually, your ideal audio level is between -10db to -20db.

The Importance of Audio Gain

In a sentence: the amplification or attenuation added to an audio signal to maintain consistent volume levels that improve the recording’s quality. Reducing the audio gain of high-level signals enhances the clarity, intelligibility, and consistency throughout an audio track. For example, multiple microphones capture audio from sources like instruments and vocalists at a live concert. But without proper gain adjustment, the vocalist could sound too quiet while the instruments sound too loud, creating an unbalanced sound mix. By adjusting the gain for each microphone, audio engineers ensure that all sources are heard at appropriate levels, creating a cohesive and enjoyable listening experience.

How to Use Audio Gain in Voice Over

You can use audio gain in voice over by following these steps:

1. Speak a few test sentences at your usual speaking volume while monitoring input levels displayed on your audio recording software or hardware.
2. If your initial soundcheck reveals that the input levels are too low, increase the audio gain on your interface or preamp until you reach the desired range. Avoid excessively high gain settings, which introduce background noise or cause distortion.
3. Record a sample with a few lines from your voice over script that represent the overall tone and delivery. Speak them naturally, ensuring you modulate your voice appropriately.
4. Listen to the recorded sample and pay attention to any issues that may arise. Check for background noises such as hums, clicks, or room echoes. Keep an ear out for any distortion or clipping that may occur during the louder parts of a recording.
5. Use audio editing software such as Audacity, Adobe Audition, or GarageBand to make precise gain adjustments during post-production. You can amplify or attenuate specific segments or the entire recording to ensure consistent volume levels.
6. Consider using audio compression during post-production to control the dynamic range of the voice-over recording.

Audio Gain Adjustment Techniques for Voice Over

Audio volume and audio gain are two distinct concepts; audio volume is how loud an audio signal is, while audio gain is how an audio signal is amplified or attenuated. Gain, however, describes how an audio signal’s strength is adjusted to make a signal louder or softer without affecting the loudness or softness a listener hears. It’s usually found in equipment like mixers, preamps, or audio interfaces.

Ultimately, that will depend on your project type and requirements. Whether it’s a podcast, film production, live event, or music album, skilled audio engineers can capture high-quality audio, maintain consistent volume levels, and address all sound challenges that hinder audio quality. They also resolve technical issues during recording or post-production, from troubleshooting audio equipment to addressing connectivity problems and utilizing advanced tools to eliminate distortion, clicks, or pops.

Best Practices for Recording

1. Run a Test First

Even if nothing has changed from the last time you did a voice over, run a test first. Top Tip: When you listen back to your test recording, use headphones to check the audio quality. While you’ll obviously want the audio to sound good on the cheapest speakers, you (and your audience) will be much happier if you use headphones to check for quality.

2. During Recording

Once you’re satisfied with your microphone placement and audio levels, you’re ready to record your voice over! Don’t stop: If you make a mistake, you don’t have to start over - you can always fix it when you edit. Take breaks: If you struggle with the script or it just gets too hard to keep going, pause your recording and take a break. Practice makes perfect: As with anything, voice overs get easier the more you do them.

3. Editing and Post-Production

When you finish recording, it’s time to edit. Even if you didn’t make any mistakes, there’s a good chance there’ll be a few things that need fixing. However, Audiate makes it so easy to edit that you’ll barely have to lift a finger. With traditional audio software, you have to hunt through your recording to find your mistakes and edit them individually. But with Audiate, you can just read the text before highlighting and deleting any mistakes you find. Does it sound like you hoped? Are there any parts that sound rushed, or too slow? Are any words unclear? Did you mumble? Once you’ve listened to your voice over, you can go back to the beginning and start editing and reducing audio noise. But remember, pauses are ok!

Audio Editing and Noise Reduction

Audio editing software is incredibly important to polish and perfect your recordings. In Camtasia, importing and working with audio only takes a couple of clicks and with Audiate, it’s even easier. You might also want to check out this post on syncing audio and video in Camtasia! Stop fearing the timeline.

A large portion of your audience will probably listen to your video via headphones. This is why adding captions and subtitles to your video, as well as providing an audio transcription, are so important. With most other audio tools, you would have to send your audio out to a professional transcription service to have an audio file transcribed and timestamped. Once you’ve recorded and edited your voice over, you can export that transcription as a caption file (and SRT file) and import it directly into Camtasia. Then, Camtasia will automatically insert the captions into your video.

Voice User Interface (VUI) Testing

Designing voice user interfaces is increasingly becoming part of user experience professionals’ job descriptions. While voice assistants and other audio-based tools mimic some of the user flow of a typical UX project, there are some unique differences.

Key Testing Techniques for VUI

1. Have a Conversation

Voice assistants aren’t yet sophisticated enough to hold a typical human conversation, where they remember details from previous conversations or understand slang. But they should be able to follow a set pattern of prompts and responses to produce the user’s desired outcome. One of the most effective ways to test this is to “perform” the script you’re creating for the application as though you’re speaking to another person. Have one volunteer read the part of the assistant while you follow the typical patterns of a human user. If you hit dead ends or get stuck in loops, you’ve spotted some areas for concern.

2. Look for Gaps in Information

When we talk to other humans, we don’t always include all the necessary information in our sentence. This is because we’re able to detect context clues and other details that an automated assistant can’t.

3. Experiment Across Different Ways of Speaking

Though voice interfaces have improved their ability to detect human speech, there are still far too many instances in which people aren’t able to use these tools due to accents or altered speech patterns. This can be embarrassing and frustrating for the end user and make it less likely that others will use the tool. To help counteract this, it’s important to incorporate different ways of speaking into the user testing process.

4. Automate Testing When Appropriate

While human user testing is critical for voice interfaces, there’s also a role for automated testing. For example, there’s no way that a human tester can replicate the millions of different variations in human speech that a voice assistant needs to handle. Similarly, human testers can’t generate every possible request or response that a voice assistant is likely to receive. For that, automated testing tools can help. Two of the most popular voice user interface testing tools are the open source tool Botium and the paid tool Bespoken.

Key Considerations for Audio Gain

Here are some frequently asked questions about gain in audio:

Question	Answer
What is a good audio gain level?	A good audio gain level is within the range of -12 to -6 dB for most audio recordings, allowing for an optimal balance between capturing sufficient signal strength and avoiding distortion or clipping.
Does gain affect sound quality?	Yes, it does because if the gain is set too high, it introduces noise, distortion, or clipping, decreasing sound quality. So, finding the right balance and avoiding excessive gain settings helps maintain optimal audio quality.
Should gain be high or low?	It depends on the specific audio source and outcome. Generally, the gain shouldn’t be too high or too low; it should be adjusted to achieve an adequate signal level without introducing noise or distortion.
What is an example of gain in audio?	This is when adjusting the gain on a microphone or audio interface to amplify the input signal from a musical instrument during a recording session. Increasing the gain makes the instrument’s sound more prominent and easier to work with during the mixing and editing stages.

Voice user interfaces will continue to evolve and improve over time.