"Servers Offering MCP Prompts as Prompt Primitives" matching MCP servers:

Enables continuous voice conversation with AI coding assistants by locally transcribing speech with Whisper and delivering utterances as text prompts.

Local MCP server that adds multimodal capabilities to text-only models like Codex/DeepSeek, offering tools for image description, audio transcription, video analysis, image/video generation, and speech synthesis.

MCP server that scores dialogue transcripts for bridging quality, offering per-speaker indicator scores with evidence and tools for parsing, comparing, and analyzing conversations.

Enables MCP-capable assistants to transcribe local audio files or URLs and perform speaker diarization for Spanish and Portuguese audio, with options for speaker count hints, domain prompts, and transcript retrieval in multiple formats.

Give your AI agent a voice with x402 pay-per-call speech synthesis, offering 20 voices, 10 personas, 31 languages, and granular controls.

Bridges MCP-capable AI tools such as Claude Code, Cursor, and Codex CLI to the Prodius desktop app, letting an agent pop a question on screen and capture your spoken or typed reply without switching back to the terminal. It also sends fire-and-forget desktop notifications and can answer permission prompts by voice.

An MCP server for audio-to-text transcription using Google's Gemini API via OpenRouter, offering multiple tools for raw, cleaned, or formatted transcripts with support for local and remote deployment.

MCP Server for automated conversational phone calls using Asterisk with Speech-to-Speech capabilities, allowing users to make phone conversations as easily as writing a prompt.

Enables reading, searching, transcribing, and sending WhatsApp messages through uazapi, with audio returned as transcribed text and media saved as local files for AI agent use.

Provides various AI capabilities through DeepInfra's OpenAI-compatible API including image generation, text processing, embeddings, speech recognition, object detection, and classification tasks. Enables users to access multiple AI models for diverse tasks like generating images from prompts, transcribing audio, analyzing text sentiment, and performing computer vision operations.

A PowerShell-based MCP server that enables Claude Desktop to convert text to speech using Windows' built-in Speech API, offering features like playback control, speed and volume adjustment.

Provides VOICEVOX text-to-speech as an MCP tool. Requires a running VOICEVOX engine on localhost.

This MCP server enables local-first video subtitle processing on Windows, offering tools for transcription, subtitle formatting, SRT/ASS export, FFmpeg rendering, and Kdenlive project creation without cloud APIs.

A Model Context Protocol server for macOS that enables AI assistants to play system sounds for audio feedback, offering informational, warning, and error sound options.

Transcribe public videos and audio (YouTube, TikTok, Instagram, Facebook, Pinterest, Google Drive) into accurate, timestamped text from Claude, ChatGPT, Cursor or any MCP client. Spanish, English and dozens of languages auto-detected. Poll-based (transcription takes minutes), same free-beta limits as the app. Docs: https://justtranscribe.ai/mcp

Enables supported local AI clients to control a visible desktop presence through bounded presentation tools such as animations, window control, and short local speech over a loopback endpoint, without granting decision or execution authority.

Enables driving a Claude Code session directly from WhatsApp over a linked device or the official Cloud API, with bidirectional messaging, media attachments, local voice transcription, per-group personalities, and context recovery after restarts. Adds pairing-code access control, allowlists and group policies, plus remote approval or denial of tool permission prompts via emoji reactions — no API keys, no Docker.

MCP server wrapping the ownvoice CLI as a single generic run tool for voice/identity checks.

Enables recording audio from a microphone and transcribing it using OpenAI's Whisper model. Works as both a standalone MCP server and a Goose AI agent extension.

Exposes the program's push-to-talk transcript history to MCP clients such as Claude Desktop and Claude Code, letting them fetch recent or latest transcripts, query results by time window, range, or ISO timestamp, search text, look up records by id, and read counts, runtime status, or clear history. Each result carries millisecond-precision hold start/end timestamps alongside the transcribed text.

An MCP server that audits voice-assistant tools for prompt injection, tool poisoning, privacy leaks and speech-unfriendly answers. It exposes scanning, PII/link checks, voice rewriting and server-audit tools so users can test whether MCP tools are safe and voice-ready.

An MCP server that provides ElevenLabs text-to-speech as a Leo provider package, enabling text synthesis to base64 audio and voice listing through the Leo speech pipeline.

Enables Claude to interact with WhatsApp: read chats, search messages, send messages with a mandatory confirmation step, and transcribe voice notes locally, all with encrypted storage and prompt-injection scrubbing.

Local voice and AI platform for macOS offering speech recognition, speech synthesis, MLX-based inference, OpenAI-compatible APIs, and MCP tools for use as an assistant or integration gateway.

Enables AI agents to ring or text a user's actual iPhone, read a question or update aloud, and return the spoken answer back to the agent as text. Calls and messages land in a single titled thread per agent, with tools for polling results, waiting for replies, and labeling conversations.

Enables Alexa+ voice intents to be evaluated and orchestrated through AWS Bedrock Claude 3.5 Sonnet, coordinating multi-surface actions across Ring, Fire TV, and Wearable AI devices via MCP tools, prompts, and resources.

Exposes Sarvam AI's speech, translation, chat, and document APIs as typed MCP tools for any MCP-capable agent.

Enables hands-free voice collaboration with a coding agent: you speak requirements aloud and hear only short, plain-language progress and completion summaries while the agent edits code, runs tests, and fixes bugs in the background. It also handles continuous speech recognition with automatic turn detection, barge-in interruption, and spoken approval prompts before risky operations.

Enables running Claude Code continuously from Telegram, with two-way voice notes, persistent memory across restarts, a self-healing daemon, and live progress showing every tool call as it happens. Also supports scheduled reminders, calendar access, inline-button prompts, message history search, and a Mini App board for monitoring sessions and token use.

Records your calls on your own computer, transcribes them on the machine, and lets an agent start, follow, question and annotate a call while it runs. No meeting bot, no account and no cloud; the same core runs as a self-hosted transcription server.