As part of Warner Music Group’s corporate social responsibility, they support initiatives in music with positive social impact. Musicians have long used their art to help with speech sound disorders (including WMG’s Ed Sheeran), so this area was a perfect fit.

Speech and language therapists have an arsenal of tools at their disposal which can help their patients. Most of these patients are children, however, which raises a problem that most tools can’t overcome - how to keep kids engaged in their therapy.

We set out to provide speech and language therapists with a unique and accessible new tool that could help their patients practise the sounds they struggle with in a way that doesn’t bore them: by reimagining speech therapy within the world of popular music.

Repetition is key to overcoming a Speech Sound Disorder, but for kids, repetition is boring. Except when it happens in music. Warner Music teamed up with Apple Music to analyse over 70 million song lyrics, isolating songs in which particular sounds occur in particular patterns that are beneficial for speech therapy. These songs were collated into Saylists: playlists categorised by problem sounds, providing an easy, accessible, and enjoyable way for kids to practise the sounds they have trouble with, simply by singing along to their favourite songs.

The strategic approach to this brief consisted of analysing an existing data set for its previously unrealised therapeutic potential. A linguistic analysis of sounds in speech therapy is not dissimilar to linguistic analysis of poetry, or song lyrics. By using one set of techniques on a unique data source, we were able to unlock therapeutic potential that had previously been untapped.

This meeting of two worlds is particularly relevant when our target audience was considered: a group which overlaps both the audience for pop music, and the age-group most likely to have a Speech Sound Disorder (SSD): young people. Saylists provides this cohort with an easily accessible way to practise the sounds they struggle with in a way that’s fun, engaging and – most importantly – not boring.

We developed an algorithm to analyse song lyrics for specific sounds occurring with a certain frequency, as well as other conditions such as proximity to similar sounds and placement within a sentence. The algorithm then analysed over 70 million songs in search of those that demonstrated genuine therapeutic value, the largest data analysis of lyrics ever. All were scored and ranked. Songs that scored highest and fit listener profile were compiled into Saylists categorised by problem sound.

The Saylists were launched on Apple Music where they are easily accessible to all, with supporting communications on social channels. The Saylists will be available indefinitely, with data available to create additional Saylists in the future to continually refresh the platform and the offering to listeners.

They’ve only been out in the world a matter of weeks, yet Saylists have already been enthusiastically welcomed and praised by people around the globe, with Spain’s El Mundo newspaper describing it as ‘Apple’s next great invention’.

Most important was the incredible response from the speech and language community. Saylists are already being recommended by therapists, and used in therapy sessions; the Royal College of Speech and Language Therapy expressed their excitement at the project’s release, and the department of English Language Teaching at Cambridge University has gone so far as to publish lesson plans based on Saylists and how they can be used in classrooms. Due to the initial response Apple Music are now planning to expand the project into other countries around the world, with playlists specific to languages beyond English. For Saylists, the story is just beginning.