# kaixxx/noScribe

Cutting edge AI technology for automated audio transcription. A nice GUI for OpenAIs Whisper and pyannote (speaker identification)

Repository: https://github.com/kaixxx/noScribe
Canonical: https://ross.abutalabs.com/products/noscribe
Language: Python
License: GPL-3.0
License Family: copyleft
Topics: audio-transcription, interview, pyannote, qualitative-research, transcription, faster-whisper
Last push: 2026-07-21T14:02:16+00:00

## Health v2 (maintenance only)
Score: 78/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 93, release rhythm 54, longevity 86
- inputs: {"age_days": 1209, "days_push": 43, "days_rel": 92, "gap_med": 237.5, "n_releases_24m": 3}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2132, forks 350 (observed 2026-08-28T04:06:17.158691+00:00)

## What it is
noScribe is a free, open-source desktop application that transcribes audio locally using OpenAI's Whisper (via faster-whisper) and pyannote for speaker identification. It includes a built-in editor for reviewing and correcting transcripts, targeting qualitative researchers and journalists.

## Use cases
- transcribe interview recordings to text
- identify different speakers in an audio recording
- transcribe audio offline without uploading to the cloud
- prepare transcripts for qualitative social research
- transcribe journalistic interviews
- review and correct AI-generated transcripts in an editor
- transcribe audio in multiple languages

## When to choose
- you need confidential, fully local transcription with no cloud services
- you need speaker diarization alongside transcription
- you work in qualitative research or journalism with interview recordings
- you want a GUI instead of command-line Whisper tools
- you need multilingual transcription support

## When to avoid
- you need fast, real-time transcription (it can take up to 3x audio duration)
- your audio quality is poor or noisy
- you cannot spare several gigabytes for AI models
- you need cloud-scale batch transcription of many files
- you want a lightweight tool without large model downloads

## Facets
- artifact type: application
- maturity: active
- function: speech-recognition, audio-processing, gui
- domain: speech-processing, developer-tools, desktop-applications
- platform: windows, cross-platform
- tags: whisper, pyannote, speaker-diarization, transcription, interviews, qualitative-research, local-ai, offline, natural-language-processing, macos, linux, desktop

## Member repositories
- kaixxx/noScribe (main) score 78

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:17.158691+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 2026-08-30T02:52:03.478218+00:00, confidence not recorded.
  - readme: https://github.com/kaixxx/noScribe (fetched 2026-08-28T04:06:17.158691+00:00, sha 312210271f13)
- Data as of 2026-08-30T08:39:29.467469+00:00.