Enables spoken conversations with Claude on a Mac: Claude speaks through speakers, listens to the user's natural replies, and transcribes them locally. No audio leaves the computer, and it includes tools for voice setup and a hands-free voice mode.
Image, video, chat and text-to-speech models (GPT Image, Gemini Image, Veo, Kling, Seedance, Claude, GPT, Gemini) behind one API key. generate_image returns a preview the model can see, and review_image has a vision model critique the result and propose a corrected prompt, so an assistant can generate, check and fix images on its own.
Enables AI agents to perform local audio tasks such as speech synthesis, voice cloning, music and sound effect generation, and audio editing through MCP, with GPU models loaded on demand and released after idle.
Enables spoken consulting case interview practice through Claude, providing real-time cases, grading on structure, analytics, judgment, and communication, and targeted practice drills.
Text-to-speech MCP server that enables AI assistants to read text aloud on the user's computer using Windows SAPI, with no API key or cloud service required.
Enables MCP-compatible AI clients to convert text into spoken audio played locally on Windows, macOS, and Linux using the Kokoro-82M model, fully offline. It also lets assistants adjust voices, speed, mute status, and auto-speak narration through a configuration tool.
Enables AI coding agents to synthesize and play concise spoken summaries of their work through multiple TTS providers, including Azure, OpenRouter, OpenAI, Gemini, ElevenLabs, and local OpenAI-compatible endpoints. It also lets agents stop playback on demand and check which provider, voice, and audio bridge are currently active.
Enables any MCP client to generate images, videos, and text-to-speech audio through OpenRouter models with a single API key, saving the results as files on disk. It also lists the available image, video, and speech models so users can choose alternatives to the defaults.
Enables MCP clients to speak text aloud on macOS using the built-in say command, with no external speech API or keys, and to list available voices and control rate or voice selection. Requests from multiple client threads share a single FIFO playback worker, so speech plays one item at a time and can be inspected, stopped, or muted in hold or discard mode.
Enables AI apps to transcribe local audio and video files on the user's own GPU, producing SRT/VTT subtitles and text without cloud uploads. It also helps find media files, list available devices, and list supported languages and models.
Renders HTML canvas animations (Canvas 2D, WebGL, three.js) to MP4 fast on a Mac, encoding H.264 in headless Chrome with WebCodecs on the hardware encoder. Gives agents 66 code-made film styles, review stills, full renders, reading-time checks, TTS and loudness mastering, with short results to save tokens.
Enables text-to-speech functionality on macOS using the say command, offering extensive control over speech parameters like voice, rate, volume, and pitch for a customizable auditory experience.
A Model Context Protocol server that integrates with VOICEVOX engine to provide text-to-speech synthesis and speaker information retrieval, allowing users to generate and play voice audio from text.
Enables agents to synthesize speech through a local OpenAI-compatible TTS service, exposing stdio tools that speak text aloud, toggle voice mode on or off, and list available voices without extra dependencies. Its settings stay synchronized with an optional sidebar panel, so one server-side switch governs all automatic narration.
An Alexa+ MCP add-on that turns a parent's description of their child's day into an original illustrated, narrated bedtime story with word-by-word read-along highlighting, branching choices, a word of the night, and a persistent story world that carries characters and past episodes from one night to the next. It runs the writing, painting and narration pipeline in the background on Amazon Bedrock and Polly while keeping every tool call under Alexa+'s latency budget, with Bedrock Guardrails checking the parent's request and every page.
Enables AI agents to transcribe audio into timestamped transcripts and subtitles, synthesize speech from text, and clone voices from authorized reference recordings, with durable asynchronous jobs and optional on-demand GPU workers.
Enables Claude Code and Codex to play spoken alerts on a Mac when a task finishes or user attention is needed, using built-in or custom voice lines while staying quiet during keyboard use, calls, Focus modes, or configured locations.
Enables agents to control a transparent, always-on-top desktop character window over a loopback-only local HTTP API — applying expressions, motions, parameters and gaze, speaking via TTS or uploaded audio with lip-sync, capturing screenshots, and switching models, window placement and visibility. Everything stays local, so no cloud account or manual token handling is required.
Exposes a single transcribe tool over streamable HTTP so containerised agents can send a media file name and receive text transcribed locally by MacWhisper on the host Mac's GPU, with token-gated access and no uploads or API keys. Callers place media in a configured directory, and the blocking call returns the finished transcript.
Enables AI agents to give spoken output by synthesizing text into speech with an on-device neural model and playing it through the user's speakers, entirely offline. Agents can queue spoken summaries and stop playback on demand.
Exposes a local, hands-free voice room to any MCP client, letting agents read speaker-labelled conversation turns, wait for the next turn, list who is present, and speak aloud through local text-to-speech. Speech recognition, speaker identification and voice output all run on the user's own machine, so agents receive text with speaker names rather than audio.
Enables local voice cloning and multi-take text-to-speech synthesis on Apple Silicon, plus audio/video transcription and subtitle generation, through Claude and other MCP clients.
Enables agents to generate on-demand text-to-speech voiceovers and receive ephemeral audio through a single tool call, using Microsoft neural voices or optional offline engines.
Enables AI agents to synthesize and play short summaries aloud while keeping full answers on screen, with queued speech and stop control using online natural voices or offline synthesis.
Enables coding agents to hold spoken conversations by exposing speech-to-text and text-to-speech tools over MCP, routing them to a local daemon that drives a Chrome window using on-device speech recognition and Piper synthesis. It supports listening, speaking, and reading files aloud within a Claude Code plugin or an MCP gateway.
Local MCP server for neural text-to-speech using Kokoro ONNX engine on CPU, supporting SSML tags, multiple voice profiles, and zero-GPU operation for low-latency speech synthesis.
Local voice toolkit over MCP: transcribe audio to text in 25 languages, synthesize speech in 9, and list available voices and languages. Runs fully on-device — no API keys, no cloud.
Enables Claude to turn a topic into a narrated, animated explainer video by generating scenes, recording them frame by frame, synthesizing speech, and returning the finished MP4. It runs on an ordinary CPU-only computer without a graphics card, paid API, or other AI service.
Generate realistic text-to-speech locally on Apple silicon with Fish Audio's Drama 3 through a stdio MCP server, including voice cloning, voice design, and directed retakes. Requires Drama Studio running on a Mac; speech generation uses Metal and Core ML.
A stdio MCP server that lets AI coding tools on a Mac speak text aloud through the local Speaklna text-to-speech app. It exposes four tools (speak, stop, status, voices) so agents can voice replies, halt playback, check state, and list available voices, but cannot import voices or change settings.
Enables Claude to discover available engines and voices, generate narration from text, monitor generation jobs, download finished audio, and run voice evaluations, all against a local on-device service.
Enables a coding agent to speak text out loud offline using an in-process text-to-speech engine on the CPU, with no API key or HTTP server. Supports seven languages, ordered non-overlapping playback across sessions, and non-blocking calls so the agent keeps working while audio plays.
Enables MCP clients to synthesize Thai text into 24 kHz WAV audio offline on CPU using MakHamTTS, optionally save it to a specified path, and play it through the default audio device.
Enables AI assistants to speak with realistic cloned voices via ElevenLabs TTS on Cloudflare Workers, supporting 29 languages and inline audio playback on both desktop and mobile.
VoiceLayer MCP server enables AI coding assistants to speak and hear via local, on-device speech-to-text and text-to-speech, with no cloud dependencies.
Voice interface for Claude Code enabling hands-free, conversational interaction entirely on-device for Apple Silicon Macs. It provides push-to-talk transcription and automatic spoken responses via local AI models.
Text-to-speech MCP server using the Kokoro-82M model accelerated with MLX on Apple Silicon, enabling local Claude and Codex clients to speak text aloud and convert text to audio.
Enables AI agents to generate narration in a single fixed, pre-approved voice, returning audio with per-line timing, hashes, and automatic quality checks on content, speaker identity, and pace, all locally and reproducibly.
Enables programmatic control of Yandex smart home from LLMs: text-to-speech on Alice speakers, batch reminders, device control, and complete scenario management.
Exposes a text-to-speech announcement tool for the CastleCall home PA system on Raspberry Pi. It enables users to broadcast voice messages using local Piper or cloud-based ElevenLabs providers through the Model Context Protocol.
A desktop avatar companion for Claude Code that provides a visible on-screen presence with text-to-speech, speech-to-text, and an interactive avatar overlay.