Podcast Transcripts Made Simple With Brightspot and Amazon Transcribe

Photo of two young people recording a podcast in a studio

Brightspot supports efficient and intuitive publishing workflows for all the content types our users produce, and podcast publishing is no exception. Why do we care about podcast publishing? Edison Research finds that approximately 90 million Americans listen to podcasts each month.

Brightspot users are now managing and producing more audio and video content on the platform than ever before to support their podcasting initiatives. With this higher volume comes more opportunities for improvement on the Brightspot platform.

We found that digital producers using Brightspot had to spend a significant amount of time or money to have their audio and video content manually transcribed. Producers use these transcriptions for a myriad of use cases, both within the CMS and on the front-end of their sites and apps.

If you’ve never transcribed manually before, there are basically two ways to achieve this:

  1. Have a living breathing person listen to content, which can range in length from two minutes in length to more than an hour. This person must then meticulously capture and write out every single word.
  2. Outsource this to a service that charges a fee (around $1 per minute) for transcribing.

If your team is regularly creating lengthy podcasts for several shows on a weekly cadence, that adds up quickly. This served as an opportunity for us to explore an integration with Amazon Transcribe.

What Is Amazon Transcribe?

So, you might be wondering, what is this service? This is how Amazon describes it:

Transcribe uses advanced machine learning technologies to recognize speech in audio files and transcribe them into text. You can use Amazon Transcribe to convert audio to text and to create applications that incorporate the content of audio files.

How Transcribe Works With Brightspot

When an audio or video file is uploaded to Brightspot, the platform can automatically send that file to be analyzed by the Amazon Transcribe service. This works for content in any of the 16 languages supported by Amazon. Brightspot then receives a response back from the service that includes the words spoken with a confidence level of the transcription. That means AWS can help identify words in the transcription it might be uncertain about so editors can clean up the transcript as needed. Transcribe can even identify if there are multiple speakers in the content.

AWS Integrations

This feature set adds up to scale: Now dozens of pieces of audio or video content can be transcribed within minutes instead of hours, and they are readily available for newsrooms to take advantage of.

The Benefits of Transcribe

Let’s talk about what a newsroom can then do with that transcription—which is plenty.

One strategy successful podcasts execute with transcripts is displaying the transcript on the front-end site of a podcast episode. This offers two benefits, first and foremost being that it improves the SEO performance of the podcast. There is a wealth of topical keywords in the audio file which you can’t expose to search algorithms and crawlers with just a headline and short description. Exposing the transcript on the site boosts SEO performance.

The second benefit is that it improves the accessibility of your site or app for audio or video content. This way your site or app’s audio/video content is accessible to more people at different levels of disability or severity of impairment.

An added benefit of transcribing audio and video files is that this content is also more search friendly within the CMS for other users. So if an editor is curating a feature story covering a state representative, they can search against all mentions of the representative in audio or video content in Brightspot, even if their name isn’t associated with the content in the headline or short description. They can then repurpose this content to create a package for their feature.

With Brightspot and Amazon Transcribe, we’re saving our users time and money.

Artificial Intelligence/Machine Learning in Brightspot’s Future

When it comes to working with AI/ML tools for your use-cases, they are just that, tools. Tools don’t simply solve your problems for you, it takes truly understanding your use-case and the problem you are trying to solve.

At Perfect Sense, understanding our customer’s use-cases and problems to provide efficient and valuable solutions is at the core of what we do. As we continue to work with our partners to explore the possibilities of AI/ML, we’ll discover even more solutions.

About the Author
Chris Smith
Chris Smith is a Platform Product Manager at Perfect Sense, where he helps build and improve the Brightspot platform. Prior to joining the company, Chris worked for CBS, where he managed the video, web development, and design team for CBS Interactive. He specializes in streaming media technology and OTT video. He is a graduate of Towson University.
bicycle for form