The StageZero speech analytics is build around an API that can be accessed using any programming language. The analytics can be setup as part of an IVR and can process live calls in real-time.
When a person calls, the AI listens to the calls an makes decisions based on what the person says.
Our analytics can be setup to process live calls or through an API to which you can send your voice clips either as streams, files, or URLs.
The data in live call can be one or multiple channels.
The API can be called either once for each recording or sent as batches of recordings. The result is returned as soon as the algorithms are done processing the request.
Our algorithms process the clips and returns a JSON formatted response with the information that was identified and suggested actions to take.
We provide a set of existing intents that you can choose to monitor for or if our existing intent do not cover your use case, we can add custom intents. To understand which one is appropriate, we need to schedule a call to understand your use case and the intents.
When an intent is recognized, it will be returned so that you can automate the action to take.
Identifying whether a person is feeling anger, surprise, disgust, enjoyment, fear, or sadness can help you take preventive actions and for example reduce customer churn. We offer emotion recognition as a popular feature of our speech API.
It may be useful to create automated actions based on identifying unwanted behavious such as swearing, profanities, or shouting. This feature is available as an add-on.
In our call center analytics suite, you can also get access to information such as: key word identification and background noise level.