Tips for Scaling Live Auto Captioning

August 14, 2020 BY SAMANTHA SAULD
Updated: March 30, 2021

two friends watching video on a smartphone

Many businesses and organizations are implementing live streams in their content strategies. Whether it’s a webinar, live after show, or virtual conference, viewers are yearning for live events more than ever before. 

With the growing popularity of online live video content, most video and social media platforms have enabled live streaming features. This is probably due to the sheer number of people tuning in to live streams. In 2019 alone, internet users watched a staggering 1.1 billion hours of live video and it doesn’t seem to be slowing down anytime soon. According to Grand View Research, the live video market is expected to surpass $184 billion by 2027. 

As the demand for live video continues to rise, many brands are looking to ramp up their streaming efforts by providing more high-quality, engaging, and accessible content. One way of doing this is by including live captions. 

Live captions are captions for events happening in real-time, as opposed to closed captions which are used for pre-recorded content. There are two main ways to generate live captions: automatic speech recognition (ASR) or a human stenographer. For the purposes of this post, we’ll focus on live auto captioning which uses ASR technology. 

If your organization is looking to scale live auto captioning for your live events, it’s imperative to find a vendor that can support your individual and unique needs. Ultimately, you’ll want to find a solution that can save you both time and money in the long run as you streamline the live captioning process. 

Watch the Toolkit for Live Captioning Events 🎥


1. Use a Top-Notch ASR Solution 🎯

Live auto captions are sometimes preferred over human-powered live captions because they minimize deletion errors in the live stream, it’s easy to automate scheduling, and it’s competitively priced. 

When scaling live auto captioning, you should consider the quality of the vendor’s ASR technology. Since live auto captioning relies heavily on the ASR engine, you’ll need a solution that uses a high-quality solution. 

Some vendors rely on their own proprietary ASR technology; many of which haven’t been considered a contender in the ASR market or widely tested externally. Therefore, it may not be a reliable option. Ensure the vendor you’re using has a reputable, state of the art ASR technology. 

You should also factor in error rate and deletion rate. When following best practices for live auto captioning quality, the accuracy rate should be a minimum of 90%. Additionally, the deletion rate should be as low as possible. This is critical to the use of ASR for creating captions, as deleting words can change the meaning and convey incorrect information. 

At 3Play Media, we use Speechmatics for our ASR – a leading speech recognition technology. When tested, it performed much better than other speech engines. 

2. Find Dedicated Support 🤝

Initially, streamlining the live captioning workflow can seem challenging. It’s possible that you may experience a few hiccups around figuring out how to set up the live captions, retrieving them after the live event, or setting up an integration. 

The vendor you choose should have a team of dedicated reps to help you every step of the way no matter how simple or complex the issue. 

When looking for a vendor, ask yourself the following questions: does the vendor have a customer support team? Do they have Account Managers for more one-on-one support? Is there support documentation that you can quickly reference at any time? These are all things to keep in mind.

3. Ensure the Ability to Upgrade After the Live Event 📝

man watches video on his phone

As mentioned above, when following best practices for live auto captioning, you can expect to achieve an accuracy rate of 90%. Live auto captions help to make your event more accessible. However, in order to ensure that they are fully accessible and compliant after the live event, vendors should allow you to easily upgrade to the full captioning process.

At 3Play, we recommend our customers upgrade to the full transcription process after the event takes place to ensure the captions are at a minimum, 99% accurate. Starting with the live auto transcript, our editors will correct any errors. The transcript will then go through another round of quality assurance. 

Many vendors don’t have the capabilities to upgrade right after or deliver a truly compliant caption file. So when looking for vendors, be sure that they can deliver fully accessible files. 

 Watch the Webinar: Toolkit for Live Captioning ➡️ 

4. Partner with an Accessibility Vendor ⚖️

When you make the decision to caption your live events, you’re opening up doors for viewers with disabilities who want equal access to online content. However, it doesn’t stop there. There are many other tools to making all of your video content, live or pre-recorded, accessible to all. 

When you partner with an accessibility vendor – which is more than a live auto captioning vendor – you can easily add other accessible tools to your videos like closed captions, audio description, translations, and more. Working with a trusted vendor that maintains quality and provides unmatched support, you can find all of your needs in one place.  

5. Use a Solution that Saves You Money 💰

When using a live auto captioning solution, it saves your organization money in comparison to using a human live captioning solution, and you will have the file available for posting after the fact. However, some vendors have different ways in which they charge customers for live auto captioning. 

There are a number of vendors that charge by the number of users. This means the more users you have, the more you’ll pay. For a large company, this can become an expensive solution since there are typically many different departments, therefore users. 

We recommend finding a solution that charges by usage instead. That means that the vendor charges by the minute. With this type of solution, organizations can have as many users as they wish without having to pay extra. This is especially useful for large enterprise companies. 

6. Look for Added User Features 🧰

man laughs at a video on his phone

Last but not least, live auto captioning is more than just captions. Sometimes, it’s the minor details about the service that can make or break your experience. That’s why we recommend investigating the user features, particularly the ones that can edify your live captioning experience. 

Does the vendor have integrations with major video streaming platforms? This is useful because once you set up the live auto captions, it streamlines and automates the workflow by posting the captions directly to the live stream. 

Can you caption webinars and meetings in certain video platforms, like Zoom? Some vendors only allow you to caption both webinars and meetings if you’re in a certain account tier. With this option, you’ll have to pay more in the long run. 

We also highly recommend a vendor that allows you to schedule the live captions in advance. This gives customers a peace of mind because their captions are set up ahead of time, leaving them space to focus on making sure their live events are the best they can be. 

Another important factor to consider is whether the vendor can handle the length of your live event. Some vendors only allow you to use their live captions for a certain amount of time. A truly valuable vendor will be able to live caption your event for an unlimited amount of time. 

In our current climate, live online video is only going to increase in popularity. As demand continues to grow, it’s important to think about live auto captions on a bigger scale rather than one-off events.

Watch the Toolkit for Live Captioning 👇

3play media logo in blue

Subscribe to the Blog Digest

Sign up to receive our blog digest and other information on this topic. You can unsubscribe anytime.

By subscribing you agree to our privacy policy.