开源免费的 Wispr Flow 替代方案 | 集成FunASR本地模型和可配置大语言模型的下一代中文桌面语音工作流

快速提取音视频内容，整理成一份结构化的markdown笔记

百聆 是一个类似GPT-4o的语音对话机器人，通过ASR+LLM+TTS实现，集成DeepSeek R1等优秀大模型，时延低至800ms，Mac等低配置也可运行，支持打断

MTools 是一个功能强大的全能桌面应用程序，集成了音视频处理、图片编辑、文本操作和编码工具，内置AI增强功能。旨在简化您的工作流程，提升生产效率

VocoType 是一款运行在本地端侧的隐私安全语音输入工具，通过快捷键即可将语音实时转换为文字并自动输入到当前应用。支持语音转文字MCP、AI 优化文本、自定义替换词典、录音视频转文字等功能，让语音输入更高效、更安全。

开箱即用的本地私有化部署语音服务，快速搭建FunASR与CosyVoice2/3后端

妙语 - 智能语音输入，妙语亦可生花。

基于 SenseVoice 的 Windows 本地语音转文字工具，支持 OpenAI 格式 API 润色，低延迟，高精度。

基于Funasr的[实时]AI语音助手

这是基于FunASR实现的区分说话人语音识别API | This is a speaker-diarization-based speech recognition API implemented using FunASR.

Realtime ASR for Desktop Subtitle

FunASR实时语音识别版，识别麦克风和电脑内播放的声音，电脑语音打字软件

基于 Qt 6 QML 开发的 AI 虚拟伴侣桌面应用,集成 ASR 语音识别、LLM 大语言模型和 TTS 语音合成,支持文字聊天、语音对话和声音克隆功能

A Python project for Chinese-to-English translations (SRT subtitles) for each episode of CCTV's "Xinwen Lianbo" (「新闻联播」), a valuable resource for language learners and researchers.

Really easy-to-use Python client for FunASR runtime server.

Really easy-to-use Typescript client for FunASR runtime server.

FunASR API is a FastAPI-based inference gateway that wraps multiple FunASR speech models behind a single HTTP surface. It manages long-running model lifecycles, exposes health endpoints for each model family, and gives you a starting point for building higher level speech services such as keyword spotting, or voice activity detection

FreeSWITCH Mod_FunASR语音识别模块，2026年最新基于此模块实现真实运营商手机号空号识别(空号检测)+关机等异常状态或早期媒体音检测,无需Asr语音识别费用。

FunASR最新语音识别模型Fun-ASR-Nano-2512实时识别热词版

基于 FastAPI 开发的 FunASR HTTP 服务，可 Docker 部署

🎙 Automated offline video transcription for macOS — FunASR + speaker diarization + language detection (zh/en/mixed). Zero cloud costs, 100% local.

极简字幕 一款本地离线运行的实时语音识别字幕软件

Local voice typing for Windows powered by SenseVoice. 15x faster than Whisper for Chinese input.

转录转录：将 B 站长视频音频转录为文本，自用为主

Real-time voice-to-text transcription with support for multiple AI models and integration with external AI services

This project presents a systematic evaluation of two state-of-the-art automatic speech recognition (ASR) models—OpenAI’s Whisper (large variant) and the FunASR (paraformer-zh) model—on a dedicated Cantonese speech dataset.

🎤 Transform speech to text on Windows with fast, local AI processing. Enjoy seamless recording and automatic integration for effective communication.

语音识别模型

Implement STT service based on FunASR

🎙️ Enable real-time speech-to-text and translation on macOS with multiple ASR backends and smooth overlay interface for seamless communication.