Speechace Pronunciation Assessment API
Description
Speechace Pronunciation Assessment API is a speech recognition service used for evaluating spoken English pronunciation, fluency, and intonation in language learning applications.
The API accepts audio recordings of learners reading scripted text or responding to open-ended prompts, then returns phoneme-level, word-level, and sentence-level scoring data. Assessment outputs include CEFR and IELTS-aligned scores, detection of mispronounced phonemes, syllable stress errors, and metrics covering fluency, rhythm, and intonation.
The service supports both scripted assessment, where the expected text is known in advance, and spontaneous speech evaluation for unscripted responses. Speechace is used by developers and publishers of language learning platforms, English as a Second Language (ESL) products, and high-stakes assessment providers to embed automated pronunciation feedback into their applications.
The API is offered through tiered subscription plans based on monthly audio processing volume, with deployment options including cloud-hosted endpoints and on-premise installations for organisations with data residency or latency requirements. Integration is performed via HTTP requests that submit audio files alongside reference text, with results returned in JSON format suitable for use in mobile apps, web platforms, and classroom assessment tools.