Yesterday we unveiled a new demonstration of high-precision timecode alignment of machine transcription + translation of Russian television news broadcasts, using a combination of Google's Speech-to-Text API and Translation API through creative repurposing of the Translation API's HTML support.

The precision alignment of this new workflow yields some eerily powerful new possibilities, with translations aligned well enough in many cases to match the back-and-forth of debates and other discussions between multiple speakers.

Stay tuned tomorrow for an early look at a sample broadcast in the Visual Explorer annotated through this new workflow!