The Washington Post is testing a news application, Truth Teller, which fact checks speeches in as close to real time as possible.
The Truth Teller was built with a combination of technologies. The prototype combines video and audio extraction with a speech-to-text technology to search a database of facts and fact checks. Truth Teller takes in video, converts the audio to text, matches that text to its database, and then displays, in real time, what’s true and what’s false.
The prototype transcribing videos uses Microsoft Audio Video indexing service (MAVIS) technology. MAVIS is a Windows Azure application which uses Deep Neural Net (DNN) based speech recognition technology to convert audio signals into words. Using this service, the Truth Teller extracts audio from videos and saves information in The Post’s Lucene search index as a transcript. Each time a video is played the fact checking starts anew. The company says the prototype needs more technical work and facts, and is tuning its algorithm.
“Do we think this can be applied to streaming video in the future?” said Cory Haik, executive producer for Digital News. “Yes. Can this work if someone is holding up a phone to record a politician in the middle of a field in Iowa? Presenting the truth is without dispute one of the most important missions of journalism. So yes, we believe it can.”
Source: The Washington Post