# jxlpzqc/TMSpeech

腾讯会议摸鱼工具

Repository: https://github.com/jxlpzqc/TMSpeech
Canonical: https://ross.abutalabs.com/products/tmspeech
Language: C#
License: MIT
License Family: permissive
Last push: 2026-07-15T08:14:09+00:00

## Health v2 (maintenance only)
Score: 82/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 92, release rhythm 69, longevity 83
- inputs: {"age_days": 1170, "days_push": 49, "days_rel": 49, "gap_med": 112.0, "n_releases_24m": 5}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1447, forks 135 (observed 2026-08-28T04:04:45.444651+00:00)

## What it is
TMSpeech is a Windows desktop application that captures system audio via WASAPI loopback and performs real-time Chinese speech-to-text, displaying results as karaoke-style subtitles. It also saves transcription history and meeting notes to files, and supports pluggable external recognizers via a command-line protocol.

## Use cases
- show real-time captions for meeting audio on windows
- transcribe online meetings to text automatically
- catch up on a meeting after zoning out by reading recognition history
- generate meeting minutes saved by date
- use a custom speech recognition model as an external recognizer
- display live subtitles even with system sound muted

## When to choose
- you need offline, low-CPU real-time Chinese speech captions on Windows
- you want to capture audio played by the computer rather than a microphone
- you want to plug in your own ASR model or command-line recognizer

## When to avoid
- you need speech recognition on macOS or Linux
- you need English or other non-Chinese transcription out of the box
- you need cloud-based, speaker-diarized transcription services

## Facets
- artifact type: application
- maturity: active
- function: speech-recognition, audio-processing, gui
- domain: speech-processing, developer-tools, windows
- platform: windows
- tags: real-time-captions, meeting-transcription, loopback-audio-capture, chinese-asr, sherpa-onnx, subtitle-overlay, desktop

## Member repositories
- jxlpzqc/TMSpeech (main) score 82

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:45.444651+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:36:06.972649+00:00, confidence not recorded.
  - readme: https://github.com/jxlpzqc/TMSpeech (fetched 2026-08-28T04:04:45.444651+00:00, sha 9cf13f828c1f)
- Data as of 2026-08-30T08:39:29.467469+00:00.