Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@mcp-speech-coach give me coaching feedback on this 30-second speech transcript"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A Model Context Protocol (MCP) server — the analysis brain of a voice speaking-coach. Exposes 15+ speech-coaching dimensions as callable tools that any host (a voice app, Claude Desktop, Orcha) can use.
Built by: Mohammed Abdul Najeeb
💡 Pairs with my VoiceCoach Lite app. The app handles the live microphone and the spoken feedback (text-to-speech); this MCP server is the measurement engine it calls. That separation is the whole point of MCP: the host runs the experience, the server supplies the capabilities.
MCP is "USB-C for AI" — a standard protocol so any AI host can call any server's tools. An MCP server cannot open a microphone or talk back on its own; it exposes tools and the host decides when to call them. So the architecture is:
Voice Coach App (host) ── live mic, TTS spoken feedback, conversation
│ calls tools over MCP
▼
mcp-speech-coach (this) ── analysis across 15+ dimensionsRelated MCP server: Clerk Chat MCP Server
Tool | Dimensions covered |
| words-per-minute · pacing · pauses · voice projection · stress/intonation |
| grammar · vocabulary richness · sentence variety · word repetition · punctuation awareness · technical-term accuracy |
| fluency/flow · filler words · confidence markers |
| pronunciation clarity (proxy) · technical-term handling |
| anxiety level · confidence level (during the session) |
| synthesizes everything into spoken-style feedback the host app can read aloud via TTS |
Design note: these are grouped by job, not split into 15 micro-tools (that would be the "too many near-identical tools" anti-pattern) and not merged into one mega-tool (that would be the vague-mode anti-pattern). Each tool is one coherent responsibility with a typed schema and structured dict output.
The app records the user (live mic)
The app transcribes + measures duration
The app calls coach_feedback(transcript, duration_s, wav_path) on this server
The server returns spoken_feedback (a short coaching script) + full metrics
The app speaks spoken_feedback aloud via TTS — that's the "voice tutor" moment
For real-time conversation, the app loops steps 1-5
pip install -r requirements.txt
python server.py # stdio transportRegister with Claude Desktop (see claude_desktop_config.example.json), then ask:
"Analyze the delivery of this transcript (duration 30s): ..." → it calls analyze_delivery.
MCP server design · multi-tool architecture · tool granularity (right-sizing) · structured output · audio + NLP analysis · separation of concerns (analysis vs. experience)
MIT.
This server cannot be deployed
Pronunciation scoring, speech-to-text, and text-to-speech for language learning
Conversational AI Coaching from calls; permissioned Team Dynamics reports in a limited U.S. pilot.
Record your pitch in Claude or ChatGPT and get instant feedback on your delivery.
Hosted speech-to-text + speech emotion/tone analysis for agents. No install; trial keys built in.
