Turn learner speech into structured feedback that a tutor, coach or voice agent can use immediately. Chivox evaluates the whole utterance and preserves the phoneme-level evidence needed for precise, helpful correction.
A useful learning experience needs to explain what happened, not merely return a pass or fail.
Use the same assessment layer across short drills, guided conversation and longer reading tasks while keeping your product experience consistent.
Chivox returns evidence. Your application decides how much feedback to show, when to retry and how an LLM should explain the result.
Imagine a B1 learner reading “I think the train leaves at three.” The sentence is understandable, but the first sound in “think” is consistently replaced with /s/. The product does not need to interrupt the whole sentence or show six scores.
“I sink the train leaves at three.”
Expected voiceless dental fricative at the start.
The remaining sounds and sentence rhythm are usable.
Pronunciation is more useful to address than pace on this turn.
“Almost there. For think, let a little air pass over the tip of your tongue: th-ink. Try just the word once, then we’ll put it back in the sentence.”
Implementation note. The score identifies the location; your lesson rules and tutor voice decide the explanation, retry length and success threshold.
Word and phoneme rows preserve the exact evidence a tutor needs to explain an error, while fluency fields help the product decide whether to coach pace, pauses or pronunciation first.
{
"overall": 85,
"pron": 88,
"fluency": { "overall": 78, "speed": 65, "pause": 2 },
"details": [{
"char": "think",
"score": 72,
"phone": [
{ "phoneme": "θ", "score": 54, "dp_type": "mispronounced" },
{ "phoneme": "ɪ", "score": 86, "dp_type": "normal" },
{ "phoneme": "ŋk", "score": 83, "dp_type": "normal" }
]
}]
}Keep the scoring detailed behind the scenes while the learner receives a clear, product-specific next step.
Yes. The product can be designed for word, sentence and longer speaking or reading tasks, with the response depth matched to the activity.
No. Most products translate the detailed response into a smaller set of age- and task-appropriate coaching cues.
Start with the learning objective, activity length and your own learner audio. A short pronunciation drill can use a tighter threshold than open speaking, while low-quality or incomplete audio should trigger a recording retry before a pronunciation correction.
Tell us what you’re building. We’ll reply within one business day with pilot credits, pricing, or a deployment plan — whichever you need first.