A microphone in front of a colorful background.
A microphone in front of a colorful background.

State of the art-ful AI: Video generation using runway.ml (part 9)

We use the text to speech features of runway.ml to create a realistic voiceover.

Welcome! This month we’re introducing one of the world’s most popular generative AI toolsets, runway.ml, and using it to generate a set of short video clips to create a promotional video for a conference. The series continues below.

Project overview

As generative AI develops at speed, we’re delving into the apps and tools currently available to help with ‘real’ creative projects. In this case, we’re using the video generation tools available at runway.ml to create short video clips to edit together to create a short promotional video for an upcoming conference. We’ll use chatGPT to suggest a script, then use runway’s text-to-speech tool to create a realistic-sounding voiceover.

Note that this is not an explainer series, though we do detail the main steps to follow at runway.ml.

In this entry in the series we use runway.ml to convert our script into a realistic-sounding voiceover.

We’ve used the text suggested by ChatGPT in the previous entry in this series as the basis for creating a ‘final’ version of the script for the video. It will need further finessing when we start to assemble the final video, but here’s the script we’ll use with runway’s ‘text to speech’ AI tool:

The Script

Welcome to TechEdCon 2024, the premier educational technology expo that’s shaping the future of learning!

Based at the brand new SuperXcel in London, England, the UKs premier expo and conference destination and easily accessible with our dedicated bus and rail stations.

Walk Europe’s largest and brightest expo halls, be inspired by thought-provoking sessions that challenge the status quo and empower yourself to redefine the educational landscape.

Network with peers to forge new partnerships and collaborate in modern spaces large and small, each designed to spark your creativity.

Experience hands-on demonstrations and exclusive reveals of the latest ed-tech tools that are about to revolutionise the learning experience.

And when it’s time to recharge, relax in our vibrant and spacious lounges, where the conversation—and the coffee—never stops!

Mark your calendars for TechEdCon 2024, Where Education Meets Innovation. Together, let’s create the classrooms of the future, today!

Using the Runway.ml text to speech tool

Here are the main steps used to convert the text into a video-ready narration!

1: Select the ‘Generate Audio’ menu to view the available audio tools. Select ‘Text to Speech’:

A screenshot of a computer.
A screenshot of runway.ml showing audio options

2: Use the ‘Select a voice’ panel to scroll between available categories of voice including ‘audiobook’, ‘british’ and ‘calm’. For each category, short audio examples are available. Below, we’ve selected the ‘Narration’ category and are testing ‘Frank’, described as ‘American, masculine’.:

A screenshot of a computer.
A screenshot of runway.ml showing audio options for text to speech

3: Paste the text into the panel at the top left. As you can see, we’ve pasted in the full script:

A screenshot of a computer.
A screenshot of runway.ml showing audio options and a script

4: Click the purple ‘Generate’ button, and wait for the conversion to complete. This feature uses runway credits. Please see their website for help with how many credits this process uses. For the script above, the tool took less than one minute to generate the speech in our chosen voice:

A screen shot of a screen with a message on it.
A screenshot of runway.ml showing a rendered audio script

5: If you’re happy with the result, click the download icon to the right of the audio player to save the file to your computer. Here’s our ‘Frank’ audio, which you can download to use in your own projects if you like:

A screenshot of a computer.
A screenshot of runway.ml showing rendered audio

6: We repeated the process using ‘Dominic’, defined as ‘Audiobook, british, masculine’. Here’s the resulting audio:

We’ll use this generated audio as we start to assemble our video in the final part of this series!

Extra features

In addition to selecting a pre-canned voice style, you can train the speech model using any specific voice, including your own! To get started, select the ‘Custom’ tab:

A screenshot of a computer.
A screenshot of runway.ml showing rendered audio

To use this feature you’ll need at least a PRO level runway account. For current monthly and yearly pricing, see the website!

A screenshot of a website.
A screenshot of runway.ml showing a paid upgrade option

Join us for the final entry in this series when we start assembling our conference video using runway’s browser-based editor!