At the upcoming IBC Show, Telos Alliance said it will demo an updated version of its Minnetonka Audio AudioTools Server file-based audio automation platform, highlighting several new AI-based workflows.
Among these new AI-based workflows is the ability to measure dialog intelligibility — “one of the most oft-cited audio-related complaints across all media platforms,” according to Telos.
If the measurements reveal that the audience will struggle to understand the dialogue easily, the audio can be extracted and processed for increased intelligibility. The company said the ability to isolate dialogue will also provide opportunities for more accurate loudness measurements.
AudioTools Server V7 will also implement an AI-driven language detection algorithm. “If the dialogue in the final audio presentation is in a language other than what is expected, AudioTools Server can drive next-step processes, such as flagging the content, alerting the operator or correcting the problem, ensuring a higher level of quality control and a better viewer experience,” said Telos in a press release.
Additional machine learning capabilities include speaker identification and diarization, clone detection to determine whether content is a genuine human voice, content detection to help identify programs, genre detection and speech-to-text generation.
“The new features in our Minnetonka Audio AudioTools Server platform, focusing on intelligent operational efficiencies and next-level automation, serve as examples of how the creative use of AI and machine learning enables broadcasters and program distributors to deliver the best possible viewing experience to their television and media audience,” said Costa Nikols, Telos’ executive strategy advisor for media and entertainment.
AudioTools Server V7 will be demonstrated at IBC in the Telos Alliance booth #8.D37, and is expected to be available in Q4 2025 through all Minnetonka Audio partners.
