Just a few weeks ago, BabbleLabs launched its Clear Cloud speech enhancement product — available as both a cloud streaming API and via a web interface for handling individual files. The early results, as we expected, have been quite surprising ;-)
Find out what I mean in this four-part blog series. We’ll post a new part every few days for over the next two weeks:
Part I: Restoring the Past
Part II: Rebuilding Speech
Part III: What’s Missing?
Part IV: Better = Cheaper
First, we’ve been struck by the uniform response from people who make video — ordinary consumers, video bloggers and professional video producers: “Wow! That’s pretty amazing.” For the amateurs, I think they are just shocked when you strip away the noise from a video — it feels as if speaker has been transported out of the noisy scene to some quiet place, yet there they are immersed in traffic, the windy South Atlantic, a club, an ice skating rink. For the professionals, they know that with great care in recording and post-production, distracting noises from field recording can be painstakingly eliminated, but it is expensive and uncertain. For them, the impact of Clear Cloud is how close it gets to professional, manual sound editing for no effort. It has become obvious to us that Clear Cloud has great potential in the video production, distribution, and sharing world.
Second, we’ve started to pick up anecdotes that suggest a whole ...