Speech-to-text tools rankings
Speech-to-text tools, ordered by GitHub stars · 23 projects
Sort
1
HandycjpaisA free, open-source, and extensible speech-to-text application that works completely offline, supporting Windows, macOS, and Linux.Rust 32.8k 2026.10.032
FluidVoicealtic-devThe fastest macOS dictation app with on-device AI enhancement, a local open-source alternative to Wispr Flow.Swift 12.0k 2026.10.133
openwhisprOpenWhisprA privacy-first cross-platform voice-to-text app with local (NVIDIA Parakeet/Whisper) and cloud models, featuring meeting transcription and AI agent interaction.JavaScript 9.6k 2026.10.194
vibethewh1teagleA privacy-oriented, local AI transcription app using OpenAI's Whisper model that transcribes audio and video on your device.TypeScript 7.7k 2026.09.055
espeak-ngespeak-ngA compact open-source speech synthesizer (TTS) supporting over 100 languages and accents, using formant synthesis method, available as a command-line tool, shared library, and SAPI5 version.C 6.9k 2024.12.126
VoiceInkBeingpaxThe best open-source alternative to Superwhisper and Wispr Flow for macOS — a no-subscription voice-to-text app with local offline models, smart context awareness, global shortcuts, and an AI assistant mode.Swift 6.7k 2026.10.017
tts-server-androidjing332TTS Server is an Android system-level text-to-speech app that bundles a Microsoft demo endpoint, supports custom HTTP requests and local TTS engines, and can tell narration from dialogue by Chinese quotation marks, ideal for reading software.Kotlin 4.5k 2023.07.268
OpenSuperWhisperStarmelA macOS real-time audio transcription app using Whisper and Parakeet models, with global keyboard shortcuts and mouse trigger support.Swift 3.0k 2026.03.039
BiBi-KeyboardBryceWGBiBi Keyboard (「说点啥」) is an AI voice input method for Android: hold the mic key and speak, with 18 ASR engines including 6 offline local models plus LLM text post-processing for fast, accurate voice input.Kotlin 844 2026.10.1610
CloneTTSsipeterCloneTTS is a lightweight, offline Android Text-to-Speech (TTS) engine enabling seamless system-wide voice cloning and high-fidelity text reading, with all inference running locally on-device. 832 2026.07.3011
Mimiyuxinomimi is a real-time system audio subtitle and translation tool for Apple silicon macOS 13+ and Windows x64, displaying live subtitles for system audio from movies, live streams, lessons, and games in a floating window, with cloud services transcribing or translating into Simplified Chinese, English, or Japanese.Rust 716 2026.10.1912
opentypelesstover0314-wAn open-source AI voice input tool, an alternative to Wispr Flow and Superwhisper, supporting macOS, Windows, and Linux.Rust 599 2026.10.1113
SayItcrosswkSayIt is an open-source voice typing tool for Windows. Press a shortcut and speak, and AI-polished text lands at your cursor. It supports local models, your own API keys, or a self-hosted backend, serving as an open-source alternative to Wispr Flow / Superwhisper.TypeScript 444 2026.10.1914
danmuaiPEPETIIA Windows desktop AI danmaku assistant that periodically captures the screen, has a vision model generate danmaku and shows them on a transparent overlay, and can also read danmaku aloud via TTS, providing an observing and commenting AI audience for streams, recordings, games and desktop content.Python 360 2026.10.0415
VocoType-linuxLeonardNJUVoCoType-linux is a high-performance offline Chinese voice input method for Linux and macOS, based on Ali FunASR, achieving ~0.1s instant text insertion, input-method-level stability, very high Chinese accuracy, and low resource usage (CPU only). It supports IBus, Fcitx5, and macOS InputMethodKit.C++ 256 2026.09.1716
MultiTTSsamitoooooMultiTTS is a multi-engine text-to-speech (TTS) application for Android, supporting multiple speech synthesis engines and offline voice packs, allowing users to configure engines, voice packs, and pronunciation rules for offline reading and system-level speech synthesis. 143 2025.08.2017
WebSpeak-client-for-TeamSpeakEchoSixHIYAWebSpeak is a self-hosted TeamSpeak 3 and 6 web client and voice gateway that lets users join voice channels from a browser while the operator keeps control of servers and access policy.TypeScript 107 2026.10.1718
ddo-ttswangz-codeddo-tts is a speech engine with built-in EdgeTTS powered by rany2/edge-tts, giving reading apps such as Legado install-and-listen text-to-speech with no extra deployment.Kotlin 81 2026.09.2119
tts-vue-nextLokerLTTS Vue Next is a desktop text-to-speech application based on Microsoft Edge TTS service, built with Vue 3, Vuetify, and Tauri, supporting real-time text-to-speech, batch conversion, and multiple audio format exports.TypeScript 71 2026.04.1820
easysubeasysub-orgEasySub is a free, fully offline Chrome/Edge browser extension that generates real-time bilingual subtitles for web audio using a local WASM speech recognition model, with no uploads, registration, or network requirement.TypeScript 63 2026.10.1521
blyyoneroomlifeBLYY (Azur Voice) is an Azur Lane shipgirl app built with Kotlin and Jetpack Compose, combining voice playback, a shipgirl gallery, secretary interaction, and several extra features.Kotlin 62 2026.10.0822
ms-edge-tts-guiJJosephphA free and open-source Windows Edge TTS GUI tool that turns articles, scripts, and subtitle texts into natural MP3 audio without requiring an API key. It offers Normal, Per-page, and Per-line working modes, with independent sentence pauses, real silence at line ends, SRT subtitles, timeline JSON, per-sentence preview highlighting, and batch ZIP export.Python 44 2026.10.1523
BaiYunGebilibilibaiyunBaiYunGe is an offline voice input software for Windows 10/11 x64 that lets you press a hotkey, speak, and release to type text. Powered by a local Qwen3-ASR-1.7B model and built-in llama.cpp inference, recognition runs entirely on-device with audio and text never leaving your machine.C# 15 2026.09.23
HandycjpaisA free, open-source, and extensible speech-to-text application that works completely offline, supporting Windows, macOS, and Linux.Rust 32.8k 2026.10.032FluidVoicealtic-devThe fastest macOS dictation app with on-device AI enhancement, a local open-source alternative to Wispr Flow.Swift 12.0k 2026.10.133openwhisprOpenWhisprA privacy-first cross-platform voice-to-text app with local (NVIDIA Parakeet/Whisper) and cloud models, featuring meeting transcription and AI agent interaction.JavaScript 9.6k 2026.10.194vibethewh1teagleA privacy-oriented, local AI transcription app using OpenAI's Whisper model that transcribes audio and video on your device.TypeScript 7.7k 2026.09.055espeak-ngespeak-ngA compact open-source speech synthesizer (TTS) supporting over 100 languages and accents, using formant synthesis method, available as a command-line tool, shared library, and SAPI5 version.C 6.9k 2024.12.126VoiceInkBeingpaxThe best open-source alternative to Superwhisper and Wispr Flow for macOS — a no-subscription voice-to-text app with local offline models, smart context awareness, global shortcuts, and an AI assistant mode.Swift 6.7k 2026.10.017tts-server-androidjing332TTS Server is an Android system-level text-to-speech app that bundles a Microsoft demo endpoint, supports custom HTTP requests and local TTS engines, and can tell narration from dialogue by Chinese quotation marks, ideal for reading software.Kotlin 4.5k 2023.07.268OpenSuperWhisperStarmelA macOS real-time audio transcription app using Whisper and Parakeet models, with global keyboard shortcuts and mouse trigger support.Swift 3.0k 2026.03.039BiBi-KeyboardBryceWGBiBi Keyboard (「说点啥」) is an AI voice input method for Android: hold the mic key and speak, with 18 ASR engines including 6 offline local models plus LLM text post-processing for fast, accurate voice input.Kotlin 844 2026.10.1610CloneTTSsipeterCloneTTS is a lightweight, offline Android Text-to-Speech (TTS) engine enabling seamless system-wide voice cloning and high-fidelity text reading, with all inference running locally on-device. 832 2026.07.3011Mimiyuxinomimi is a real-time system audio subtitle and translation tool for Apple silicon macOS 13+ and Windows x64, displaying live subtitles for system audio from movies, live streams, lessons, and games in a floating window, with cloud services transcribing or translating into Simplified Chinese, English, or Japanese.Rust 716 2026.10.1912opentypelesstover0314-wAn open-source AI voice input tool, an alternative to Wispr Flow and Superwhisper, supporting macOS, Windows, and Linux.Rust 599 2026.10.1113SayItcrosswkSayIt is an open-source voice typing tool for Windows. Press a shortcut and speak, and AI-polished text lands at your cursor. It supports local models, your own API keys, or a self-hosted backend, serving as an open-source alternative to Wispr Flow / Superwhisper.TypeScript 444 2026.10.1914danmuaiPEPETIIA Windows desktop AI danmaku assistant that periodically captures the screen, has a vision model generate danmaku and shows them on a transparent overlay, and can also read danmaku aloud via TTS, providing an observing and commenting AI audience for streams, recordings, games and desktop content.Python 360 2026.10.0415VocoType-linuxLeonardNJUVoCoType-linux is a high-performance offline Chinese voice input method for Linux and macOS, based on Ali FunASR, achieving ~0.1s instant text insertion, input-method-level stability, very high Chinese accuracy, and low resource usage (CPU only). It supports IBus, Fcitx5, and macOS InputMethodKit.C++ 256 2026.09.1716MultiTTSsamitoooooMultiTTS is a multi-engine text-to-speech (TTS) application for Android, supporting multiple speech synthesis engines and offline voice packs, allowing users to configure engines, voice packs, and pronunciation rules for offline reading and system-level speech synthesis. 143 2025.08.2017WebSpeak-client-for-TeamSpeakEchoSixHIYAWebSpeak is a self-hosted TeamSpeak 3 and 6 web client and voice gateway that lets users join voice channels from a browser while the operator keeps control of servers and access policy.TypeScript 107 2026.10.1718ddo-ttswangz-codeddo-tts is a speech engine with built-in EdgeTTS powered by rany2/edge-tts, giving reading apps such as Legado install-and-listen text-to-speech with no extra deployment.Kotlin 81 2026.09.2119tts-vue-nextLokerLTTS Vue Next is a desktop text-to-speech application based on Microsoft Edge TTS service, built with Vue 3, Vuetify, and Tauri, supporting real-time text-to-speech, batch conversion, and multiple audio format exports.TypeScript 71 2026.04.1820easysubeasysub-orgEasySub is a free, fully offline Chrome/Edge browser extension that generates real-time bilingual subtitles for web audio using a local WASM speech recognition model, with no uploads, registration, or network requirement.TypeScript 63 2026.10.1521blyyoneroomlifeBLYY (Azur Voice) is an Azur Lane shipgirl app built with Kotlin and Jetpack Compose, combining voice playback, a shipgirl gallery, secretary interaction, and several extra features.Kotlin 62 2026.10.0822ms-edge-tts-guiJJosephphA free and open-source Windows Edge TTS GUI tool that turns articles, scripts, and subtitle texts into natural MP3 audio without requiring an API key. It offers Normal, Per-page, and Per-line working modes, with independent sentence pauses, real silence at line ends, SRT subtitles, timeline JSON, per-sentence preview highlighting, and batch ZIP export.Python 44 2026.10.1523BaiYunGebilibilibaiyunBaiYunGe is an offline voice input software for Windows 10/11 x64 that lets you press a hotkey, speak, and release to type text. Powered by a local Qwen3-ASR-1.7B model and built-in llama.cpp inference, recognition runs entirely on-device with audio and text never leaving your machine.C# 15 2026.09.23