Google has this week launched a new speech enhancement option within YouTube Stories which will enable users to reduce background noise in their video clips, and put more emphasis on the main subject.
As you can see (or hear) in this video, the new option, which users can apply in their upload settings, reduces background sound, improving the main presentation and clarifying speech.
The technology is part of Google’s ongoing work on speech-to-text translation, utilizing machine learning technology to isolate the speech of a video’s subject.
As explained by Google:
“By training the model on a large-scale collection of online videos, we are able to capture correlations between speech and visual signals, such as mouth movements and facial expressions, which can then be used to separate the speech of one person in a video from another, or to separate speech from background sounds. This technology not only achieves state-of-the-art results in speech separation and enhancement (a noticeable 1.5dB improvement over audio-only models), but in particular, can improve the results over audio-only processing when there are multiple people speaking, as the visual cues in the video help determine who is saying what.”
That obviously has applications beyond improving your Stories clips, but YouTube Stories provides a good testing ground for the process, while also giving users another way to improve their Stories presentation.
To apply the new option, YouTube creators who are eligible for YouTube Stories can record a video, and then select the “Enhance speech” option from the volume controls editing tool (iOS only at this stage).
Of course, not everyone has access to YouTube Stories creation as yet. YouTube Stories is only available to channels with over 10,000 subscribers, though the platform is looking to open it up to more users sometime soon.
Interestingly, YouTube will also soon stop YouTube Stories creators from adding music from the Audio Library to their Stories clips – which seems like a step back, but YouTube says the option sees very low usage.
Regardless of availability, it’s an interesting new option, and it’ll also be interesting to see whether Google looks to make its background audio reduction tools available in more places in future.
You can read more about the technical process behind the new option here.