Local audio/video transcription with speaker diarization and live audio support. No API keys. Powered by faster-whisper.
本地语音唤醒:常听麦克风 + DSH 内置 SenseVoice 识别;只说唤醒词,听到提示音后说内容,自动发送到会话;带可拖动 HUD 卡片。
Official JavaScript SDK for the Varmir speech-to-text API: transcription, translation and live streaming (Node 22+; browsers via stream tokens).
Pre-model audio processing adapters for Assembly Line agents.
Fully local, on-device speech-to-text Transcriber for moxxy. Runs multilingual Whisper or Parakeet (English + Polish) via sherpa-onnx in a forked sidecar process; the chosen model downloads once on first use, pinned and sha256-verified. No API key, no net
React component for real-time audio recording with AI transcription - works out of the box with embedded workers
DODO — local-first MCP server with multi-project coding, AI providers, sub-agents, context, memory, media and workflow tools; embedded OAuth and workspace permissions.
Local speech-to-text for web apps. Add private Whisper-powered dictation to any input with no backend or API key.
A sample browser app for Bluemix that use the speech-to-text service, fetching a token via Node.js
Whisper speech recognition
贾维斯式语音助手 for DSH:本地唤醒词「贾维斯」+ 中文识别自动填入并发送 + 语音朗读回答,Windows 专用。
Turn any book into an audiobook in DSH: search public-domain libraries or import EPUB / PDF / TXT, listen, download as MP3 with subtitles, voice-note your thoughts and discuss with AI. · 搜书听:搜书、听书、转 MP3 下载、语音记心得。
Local Russian dictation with English technical terms and a durable personal glossary for DeepSeek Harness.
Metaso ASR for DeepSeek Harness: direct Metaso login, account hotwords, and a click-to-talk composer microphone with live streaming transcription. No local service or extra speech bundles.
AI provider primitives — thin SDK wrappers for OpenAI, Gemini (image, Omni video, music, TTS, STT), ElevenLabs, HeyGen, Seedance, Fal, Kling and stock/CC media search (Pexels, Pixabay, Giphy, Tenor, Wikimedia, Openverse). Pure I/O (no DB, no runtime), ESM
Voice cloning, text-to-speech and speech-to-text for DeepSeek Harness, powered by MiniMax · DeepSeek Harness 的 MiniMax 音色复刻 / 文本转音频 / 语音转文字插件
The provider-neutral speech-to-text port: the `SttAdapter` contract and neutral transcript types, the adapter registry, the media-resolution seam, the shared transcript normalizer, and a deterministic fake adapter for tests. Concrete providers (e.g. `@ray
The speech to text API provides two endpoints, transcriptions and translations, based on our state-of-the-art open source large-v2 Whisper model.
A speech to text module for the discord voice implementation
Twick cloud function for generating JSON captions from audio using Google Cloud Speech-to-Text
Extension for making discord speech recognition bots.
Cloudflare Workers AI multimedia tools for DeepSeek Harness: text-to-image, text-to-speech and speech-to-text on the free Workers AI plan, with a settings page where each capability's default model is chosen.
Universal video downloader for HLS, DASH, YouTube and other supported sources.
On-device speech-to-text and voice control for web applications with Moonshine.
听懂:把英文有声读物(MP3)或英文文档变成听力课,练到「一听就懂」。英文音频 → 带时间的字幕 → 时间对齐的中文配音;文档 / 文字 → AI 朗读音频 → 下载到本地。DeepSeek Harness (dsh) 插件。
A Capacitor plugin that provides offline speech-to-text functionality for Android and iOS platforms. The plugin offers true offline recognition for Android with multiple languages, while iOS provides offline support for English with online fallback for ot
Detects wake words and transcribes live speech with microphone state events.
A powerful React hook for real-time voice streaming, designed for AI-powered applications. Perfect for real-time transcription, voice assistants, and audio processing with features like silence detection and configurable audio processing.
Speech-to-text and text-to-speech for OpenCode. Record voice prompts with whisper-cpp, hear responses via Piper TTS, with LLM normalization through any OpenAI-compatible endpoint.
MCP server for typeheard.com: transcribe audio and video files from any MCP client.
MCP server for Scriptivox AI transcription — transcribe audio/video from any AI assistant. 119 languages, speaker diarization, word-level timestamps, full CRUD on transcriptions (submit, cancel, delete, list), and SRT/VTT/text export.
Press-to-talk dictation: a key combination records in the DSH Web page, DSH's own speech engine transcribes, and the text lands in the composer.
React Native Speech To Text for android
A lightweight library for integrating speech-to-text functionality into your React applications, using the Web Speech API under the hood.
Deepgram speech-to-text adapter for @absolutejs/voice
MCP server for Gladia pre-recorded transcription and live job management
RecFn Whisper ASR adapter — local offline transcription via whisper.cpp
Read-aloud, dictation and language-aware speech for the DeepSeek Harness web UI: many TTS/STT providers at once (EU vendors first: Mistral, Speechmatics, Gladia, ReadSpeaker; plus ElevenLabs, Azure, Google, Amazon, OpenAI and OpenAI-compatible servers, De
RecFn Google Speech-to-Text ASR adapter — cloud transcription via Google Cloud
Better input experience for DeepSeek Harness: voice input with AI polishing, prompt optimization, and a local prompt template library
Desktop blue fish pet for DeepSeek Harness: per-pixel transparent WinForms window, text chat, and cloud speech-to-text voice conversation.
Local speech recognition with NVIDIA Parakeet TDT 0.6B v3 on the CPU: OpenAI-compatible transcription API, sendTo and a live test page
Effect AI provider for ElevenLabs — SpeechModel (TTS, one-shot + streaming), TranscriptionModel (Scribe STT), and VoiceConversionModel (speech-to-speech).
MCP server for real-time audio transcription using OpenAI Whisper
Hold-to-talk voice input for the DeepSeek Harness Web composer: hold the mouse on the input box, speak, see the running transcript float above it, release to insert the text into the draft. Recognition runs locally (SenseVoice via sherpa-onnx) — no API ke
Advanced speech-to-text transcription tool using OpenAI Whisper with GPU acceleration support