pengzhendong/streaming-sensevoice

Pseudo Streaming SenseVoice with Hotwords

What it solves

It enables real-time, streaming audio transcription by processing inference in chunks using the SenseVoice model. This removes the need to wait for an entire audio file to be processed before receiving text output.

How it works

The project implements a streaming wrapper around the SenseVoice model. It provides three ways to handle audio: processing pre-recorded wav files, capturing audio directly from a microphone, and utilizing a WebSocket service (built with FastAPI) to transmit audio chunks in MP3 format from a browser to a server for low-latency transcription.

Who it’s for

Developers looking to integrate real-time speech-to-text capabilities into their applications using the SenseVoice model.

Highlights

  • Real-time transcription from microphone input.
  • WebSocket-based server for browser-based audio streaming.
  • Support for wav file transcription.
  • Integration with the Recorder library for efficient audio transmission.

Related

  • Project
    xpzouying/x-mcp

    X‑MCP is a Chrome extension that lets AI agents (Claude, Codex, etc.) control your browser to create and publish posts on Xiaohongshu via the Model Context Protocol (MCP). It requires no Python/Docker backend—just a token from aredink.com—making AI‑driven automation reachable for non‑technical creators.

  • Project
    Wing900/ManimCat

    ManimCat is an AI‑assisted web workspace that turns natural‑language math prompts into animated (Manim) or static (matplotlib) visuals. It offers a quick‑generation “Workflow” mode and a longer‑lived “Agent/Studio” mode for iterative creation, built with React, Express, Redis, and optional Supabase storage, and can be run locally or via Docker/Hugging Face Spaces.

  • Project
    WecomTeam/wecom-openclaw-plugin

    A WeCom channel plugin for the OpenClaw AI‑agent platform. It adds Bot (WebSocket or JSON webhook) and Agent (encrypted XML webhook) connectivity, supports rich media, proactive and streaming replies, fine‑grained access control, multi‑account setups, and exposes dozens of WeCom business APIs (messaging, email, docs, spreadsheets, calendar, meetings, file storage, contacts) via a built‑in `wecom‑cli` tool.

  • Project
    ttlequals0/MinusPod

    MinusPod is a self‑hosted Docker service that transcribes podcasts, uses an LLM to spot ad segments, cuts them out, and serves ad‑free feeds. It supports local Whisper‑based transcription, multiple LLM providers, pattern learning from user corrections, and a web UI for review.