If you are like most people, you insert filler words – ‘ums’ and ‘uhs’ – into conversations when you’re grasping for what to say next. Filler words can be annoying to listen to if there’s a lot of them, and they can also add unnecessary time to podcast recordings. Until now, podcasters have had to either leave filler words in recordings or tediously edit them out.
The podcast recording, transcribing, editing, and mixing tool, Descript, just released a new version of its app that now transcribes and removes filler words from the audio recording. Descript is leveraging its proprietary Lyrebird AI to to detect and seamlessly remove the ‘ums’ and ‘uhs’ from the recordings.
Descript uses Google Speech for automated transcriptions and has also considered using Rev’s automated transcription service. However, they use their technology, Lyrebird AI, to modify the audio recording. Lyrebird AI doesn’t just remove audio, it can also replace words using the speaker’s voice. Here’s an example of it in action. 🔊 Requires sound
The filler removal tool is a feature that sets Descript apart from its competition. If Descript continues to leverage the Lyrebird AI in innovative ways, they should experience exponential growth and position themselves well for acquisition by a company like Google or Microsoft.