MetadataIQ for News uses news-specific speech-to-text technology and machine learning capabilities to describe quickly the assets that appear in a news broadcaster’s feed. As a result, editors and producers can find what they need and create news shows more efficiently.

News providers get multiple news feeds from bureaus and other providers, such as The Associated Press. There can be as many as 100 continuous feeds at a time, depending on the size of the news organization. It is impossible for humans to keep up with tagging all that media. There are many speech-to-text engines on the market that can produce decent output on general-purpose language, but they tend to falter when it comes to identifying names and places that are in the news.

Unlike general-purpose STT engines, MetadataIQ for News was made just to handle news content. It ingests all feeds in real time and applies a speech-to-text engine that is fine-tuned for news production. Digital Nirvana uses its MonitorIQ automated broadcast monitoring application to ensure MetadataIQ for News continuously improves. Every day, MonitorIQ records multiple news channels, including the audio, which contains the proper pronunciation. Then, using AI, MonitorIQ automatically identifies all names and places, and feeds all the data back into the STT engine's dynamically updated vocabulary. This process is what makes MetadataIQ for News unique and able to generate highly accurate output.

Two major news organizations — one in the United States and another in the Middle East — have been testing these capabilities in their live news operations. They have found that applying ML technologies to their workflows yields automatically generated, highly accurate metadata more quickly and affordably than traditional methods, which translates to money, effort, and time savings. It also means better-structured, more detailed, and more accurate metadata and shorter content delivery cycles.