Soniox Launches Omnio AI Model

Soniox has launched Omnio, a multimodal artificial intelligence product capable of natively understanding and reasoning speech and audio.

Omnio is a major breakthrough in voice AI because, unlike other AI models that have to convert audio into text tokens to then determine meaning and respond, Omnio not only identifies speakers and provides a transcript but analyzes tone and extracts key quotes when speakers expressed uncertainty or excitement and summarizes their tone throughout the calls.

Omnio excels at identifying speakers, their roles, and even the nuances of their interactions, including emotions, sentiment, and speaking styles. Beyond words, Omnio also recognizes sounds and non-verbal cues. It directly processes the audio signal and has been trained to recognize and understand foundational audio and speech concepts like humans.

In addition to processing audio, Omnio is also an AI model for text reasoning.

"We believe Omnio marks a significant step forward in achieving general speech and audio intelligence," Soniox said in a blog post last week.