11% of news articles contain embedded tweets

Patrick O'Flaherty
Jul 29 · 2 min read

At SpeechKit we’re always developing better ways to help news publishers bring audio versions of their news articles to readers using text-to-speech.

For news publishers, using Google Cloud TTS or Amazon Polly to convert news articles into audio has its challenges.

We developed NewsNet, a middle-layer that processes article content prior to speech synthesis, to overcome the challenges that publishers face with Google Cloud TTS and Amazon Polly.

One issue that is particularly tough for text-to-speech, when it comes to news content, is embedded tweets. With more political discourse taking place on Twitter, journalists are increasingly citing Tweets in their articles using embedded tweets.

Embedded Tweet

The problem with audio versions of news articles generated using text-to-speech is that embedded tweets often read like this:

TTS without NewsNet

To overcome this issue we added a new feature to NewsNet that processes embedded tweets prior to speech synthesis so that they read like this:

TTS with NewsNet

It’s a small feature, but with 11% of news articles now containing embedded tweets and more people using audio to listen to news articles, the impact on user experience when it comes listening is significant.

For more information on NewsNet, or how we can help you bring audio to your news articles, email us at hello@speechkit.io.


The best way to make your news stories available in audio.

Patrick O'Flaherty

Written by

Co-Founder & CEO SpeechKit



The best way to make your news stories available in audio.

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade