Reference

Speech Scoring Glossary

These are the terms behind pronunciation-scoring APIs — the algorithms, the processing steps, and the evaluation vocabulary that vendors use in their documentation and that determine what a scoring product can and cannot actually do. Each entry states what the thing is in one paragraph, then explains how it works and where it breaks.

The terms

How to use this glossary

If you are choosing between pronunciation-scoring vendors, read Goodness of Pronunciation first — almost every commercial engine in this category is a GOP variant, so its limitations are the industry's limitations rather than any one product's. Then read mispronunciation detection, which explains why a single published accuracy percentage is close to uninterpretable without the base rate of errors in the test set.

Everything here is written to be read on its own. Where a claim comes from our own evaluation runs, the page says so and gives the sample size.

Glossary questions

What is a pronunciation scoring API?
A pronunciation scoring API takes an audio recording plus the text the speaker was asked to say, and returns a measure of how closely the production matched the expected pronunciation. Most return a score for the whole utterance, for each word, and for each individual phoneme within those words.
Which terms do I need to understand before evaluating one?
Four carry most of the weight: Goodness of Pronunciation, the algorithm that produces the per-phoneme number; forced alignment, the step that decides which milliseconds of audio each phoneme occupies; phoneme scoring, the granularity at which results are reported; and mispronunciation detection, the task of deciding which sounds were actually wrong. Each has its own page here.
Why does a vendor glossary talk about limitations?
Because the limitations are the part that determines whether a product works. Every term on this list has a well-documented failure mode — score saturation, alignment error, invisibility of distortions, misleading accuracy metrics — and a buyer who does not know about them will be surprised in production rather than in procurement.