Speech AI Examples

You can access production-ready MCP servers for Brainiall Speech AI APIs, including Pronunciation Assessment, Speech-to-Text, and Text-to-Speech. These HTTP endpoints let you integrate pronunciation scoring, transcription with timestamps, and natural-sounding speech synthesis into your AI agents and IDEs with simple, scalable calls that run on CPU.

How to use

You will connect your MCP client to the available HTTP MCP endpoints to perform pronunciation assessment, transcribe audio, and synthesize speech from text. Each endpoint accepts audio input and returns structured results you can feed into your AI agent or educational tool. Use the following pattern to start a session and then call the endpoints in sequence as needed for your workflow.

How to install

Prerequisites: You need a client capable of making HTTP requests. You can use Python, Node.js, or curl based on your environment.

1. Obtain an API key or subscription for the Speech AI services.

2. Set your API key in your environment.

export SPEECH_AI_API_KEY="your-subscription-key"

Quick start examples

You can run examples in Python, Node.js, or with curl to exercise all three APIs in a single flow.

Python example installs dependencies and runs the basic usage script.

Node.js example runs with Node.js 18+.

Curl example executes ready-made commands.

Available tools

Pronunciation Assessment

Scores pronunciation accuracy at the word and phoneme level for spoken input.

Speech-to-Text

Transcribes audio and provides word-level timestamps and confidence scores.

Text-to-Speech

Generates natural speech from text using multiple English voices.