Closed syama-aot closed 2 months ago
You need to define these thresholds in your own code, instead of passing them via setting or config.
We just return these values in output json: UnexpectedBreak - Confidence and MissingBreak - Confidence. Developers can compare these values with their defined thresholds to get the final decision on error types: unexpected or missing break.
BTW, 0.75 is a default value we recommend, better to adjust them based on your own scenario.
@yaden2018 question: are these thresholds only returned when using json type? I don't see the confidence scores defined in SDK object: https://learn.microsoft.com/en-us/objectivec/cognitive-services/speech/spxpronunciationassessmentresult (I'm using Swift SDK).
Thank you!
You can find these scores in words part. If you can't get such info, please open a new issue if you are still interested in using this SDK. I'm not expert on SDK related issues.
For The following table describes the prosody assessment results in more detail:
Reference : https://learn.microsoft.com/en-us/azure/ai-services/speech-service/how-to-pronunciation-assessment?pivots=programming-language-javascript
Thresholds | Suggested thresholds on both confidence scores are 0.75. That means, if the value of UnexpectedBreak – Confidence is larger than 0.75, it has an unexpected break. If the value of MissingBreak – confidence is larger than 0.75, it has a missing break. If you want to have variable detection sensitivity on these two breaks, you can assign different thresholds to the UnexpectedBreak - Confidence and MissingBreak - Confidence fields.
I'm looking for a way to adjust these threshold values depending on the country, but I haven't found any options or settings for that.