Production speech AI server with pronunciation scoring, speech-to-text, and text-to-speech. 10 tools, 7 resources, 3 prompts for language learning and speech assessment workflows. Powered by a 17MB model that achieves sub-300ms latency with phoneme-level accuracy.
Server Config
{
"mcpServers": {
"speech-ai": {
"url": "https://pronunciation-mcp.thankfulfield-a7857897.eastus.azurecontainerapps.io/mcp",
"headers": {
"Content-Type": "application/json"
}
}
}
}