Инструменты

OpenAI's open-source general-purpose speech recognition model. The whisper command transcribes many languages, translates foreign speech into English and writes subtitle files. Needs FFmpeg; faster with a GPU.

C/C++ port of OpenAI's Whisper speech recognition model that transcribes audio offline and outputs subtitles. Runs on CPU, with Apple Silicon and various GPU acceleration. Download a ggml-format model first.

Continuously records your screen and audio locally, extracting on-screen text and transcribing speech, so agents can search pages, docs and meetings you saw and summarize your day. Data stays local. Needs screenpipe running: source is free for personal use, desktop app is a subscription.

Industrial-grade open-source speech recognition toolkit from Alibaba's ModelScope, with an agent-friendly funasr command that turns audio into text, JSON or SRT subtitles in one step, with speaker diarization and timestamps. Strong on Chinese. Requires PyTorch.

Official MCP from the FunASR project: transcribes audio files locally with the SenseVoice model, supporting Mandarin, Cantonese, English, Japanese and Korean, with optional segment timestamps. Local inference needs no API key and audio stays on your machine. Docker image available.

Unofficial Python library and CLI for driving Google NotebookLM (now Gemini Notebook): bulk-import sources, ask cited questions, and generate and download podcasts, videos, slides, quizzes and flashcards. Ships an agent skill. Relies on undocumented APIs and needs a Google sign-in.

Open-source zero-shot TTS via f5-tts_infer-cli: from a few seconds of reference audio it reads Chinese or English text in the same voice, handy for lessons and videos. Needs Python, PyTorch and FFmpeg, ideally a GPU. Pretrained models are non-commercial; get consent before cloning a voice.

Rewrites an article or script as narration, then builds a click-through 16:9 web presentation you can screen-record into an explainer video, with optional synthesized voiceover. Good for lessons, YouTube tutorials and product walkthroughs. Needs Node.js; voiceover needs a MiniMax or OpenAI key.

Converts text to speech with Microsoft Edge's online TTS service and can generate subtitle files at the same time. No API key needed, with voices for many languages. Not an official Microsoft project and depends on that online service staying available.

Offline speech toolkit for Apple Silicon Macs: CLI text-to-speech (including Chinese and voice cloning) and speech-to-text, for voicing lessons and explainer videos or transcribing lectures and interviews, with data staying local. Needs an M-series Mac and Python 3.10+; MP3 export needs ffmpeg.

One command extracts or summarizes web pages, PDFs, YouTube videos, podcasts and media files, with JSON output. Summaries can use many model APIs, local models or a logged-in Codex or Claude CLI; plain extraction needs no key. Also has a browser sidebar extension.

Open-source alternative to NotebookLM's podcast feature: turns web pages, PDFs, images, YouTube videos or text into two-host conversational audio, useful for lecture notes or articles. Uses Gemini and OpenAI API keys by default, or local models plus free Edge TTS.

Unofficial nlm CLI and MCP server for NotebookLM (now Gemini Notebook): bulk-create notebooks, import web pages and docs, and generate audio overviews, quizzes, flashcards, mind maps and slides for teaching and study. Relies on undocumented APIs and a Google sign-in, so it may break.

Unofficial CLI and MCP that let agents work with NotebookLM (now Gemini Notebook): bulk-import sources, ask questions with citations, and generate audio overviews, quizzes, flashcards, mind maps and slides. Uses undocumented APIs and may break; needs a Google sign-in and uv.

Local neural text-to-speech engine that converts text to speech offline, with voices in dozens of languages, handy for listening materials or narrating articles and course slides. Needs Python; download a voice model before first use. The project is currently looking for maintainers.

Gemini API development skills: gemini-api-dev (core API and SDKs), gemini-live-api-dev (real-time bidirectional audio/video streaming) and gemini-omni-flash-api (Omni video generation and editing), filling in APIs released after model training cutoffs. Maintained by Google's Gemini team.

Turns a topic or article into a narrated explainer video in English or Chinese: research, script, voiceover, storyboard and per-shot animation, with subtitles and chapter cards. Needs Node.js, ffmpeg and Python; takes about 1-3 hours depending on length. Free for non-commercial use only.

MiniMax's official CLI, built for agents: chat with MiniMax models from the terminal and generate images, video and speech, plus speech transcription, image understanding and web search. Works with the international platform. Requires a MiniMax Token Plan subscription.

Local text-to-speech CLI built on the open Kokoro model: batch-converts TXT, EPUB and PDF to MP3 or WAV by chapter, with multiple languages, voice blending and speed control; handy for audio versions of course materials. Needs Python 3.11-3.12 and two model files downloaded first.

Audio plugin suite: transcription with local Qwen3-ASR or Whisper (MLX on Apple Silicon), speaker diarization, transcript correction, structured meeting notes and StepFun TTS. StepFun ASR and TTS need an API key.

Turns ComfyUI workflows into MCP tools with no code, covering generation and conversion across text, image, speech and video; works with local ComfyUI or RunningHub in the cloud. Start the server with uvx pixelle and configure an LLM first.

Calls image and video models such as FLUX, Veo and Seedance through inference.sh's belt CLI to generate images, video, avatars and voiceovers, plus guides for thumbnails, storyboards, press releases and social copy. Needs belt and a free inference.sh account; model calls are billed per use.

A macOS say-style text-to-speech CLI that uses ElevenLabs or 60db to speak text aloud or save it as audio (ElevenLabs supports streaming playback), and can list and switch voices. Needs an API key for the chosen service.

Voice skills: TTS with subtitle timing, emotional character voices, translated dubbing, two-host news podcasts, sound effects and songs. Narration and podcasts work with local Kokoro or free voices; emotion, cloning, dubbing, effects and songs need a Noiz API key. Never clone voices to deceive.