Answers to the first six questions every team asks.
Integration speed, language coverage, clients, streaming, accuracy, pricing. If yours is not here, the contact form below is the fastest way to reach the team.
/Integration
2 answersHow fast can I integrate?
Minutes. Drop one object into your MCP client config, set the API key, and your agent can call `assess_speech` as a tool. No SDK wrappers, no ML setup.
Which MCP clients work?
Cursor, Claude Desktop, Cline, Windsurf, Zed, and any other MCP-compatible client. Also works as a tool inside LangChain, LlamaIndex and the OpenAI Agents SDK via the MCP adapter.
/Capability
2 answersWhich languages are supported?
Mandarin Chinese and English are first-class, both with phoneme-level scoring. Chinese includes dedicated handling for tones, pinyin, neutral tone, erhua and tone sandhi. English includes CEFR-aligned scoring with stress and rhythm diagnostics.
Can I stream audio in real time?
Yes. A WebSocket streaming session accepts mic audio frames and returns scores within a few hundred milliseconds of end-of-speech. File evaluation supports mp3 / wav / m4a / ogg / aac / pcm.
/Trust
1 answerHow accurate is the scoring?
The underlying engine has 95%+ correlation with human expert rubrics, validated by national standardized tests used across 100+ cities, with 9.2B+ evaluations per year.
/Commercial
1 answerWhat does it cost?
Free credits on signup. Tiered pricing scales with usage — higher volumes get lower unit prices. Contact sales for enterprise SLAs.
Let’s build your voice agent together.
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.