MecoScribe

Local CLI for diarized transcription on macOS, built on FluidAudio. Upload an audio file and get a speaker-labeled transcript plus an interactive HTML viewer.

Features

• Diarized transcription — identifies speakers and assigns words to each speaker
• Plain-text output — <filename>.txt with timestamps and speaker labels
• Interactive HTML — <filename>.html with:
  • Color-coded speaker segments
  • Word-level highlighting synced to playback
  • Click any word or segment to seek in the audio
  • Rename speakers
  • Built-in audio player (with playback speed selection)

Requirements

• macOS 14+
• Swift 6.0+
• Apple Silicon recommended (uses CoreML / ANE via FluidAudio)
• Network access on first run (models download from Hugging Face)

Install

git clone https://github.com/HeyMeco/MecoScribe.git
cd MecoScribe
swift build -c release

The binary is at .build/release/mecoscribe.

Model cache

Downloaded FluidAudio models are stored in ./models by default (relative to your current working directory). Subsequent runs reuse the cache instead of re-downloading.

Override with --models-dir /path/to/models or the MECOSCRIBE_MODELS_DIR environment variable.

Usage

# Basic — writes meeting.txt and meeting.html next to the audio file
swift run mecoscribe meeting.wav

# Specify output directory
swift run mecoscribe interview.mp3 --output-dir ./transcripts

# Offline diarization (default) — multilingual ASR is used automatically
swift run mecoscribe call.m4a --output-dir ./transcripts

# English-only ASR (optional)
swift run mecoscribe english.wav --model-version v2

# Preset speaker names
swift run mecoscribe panel.wav --speakers "Alice,Bob,Carol"

Options

Flag	Description
-o, --output-dir <dir>	Output directory (default: same folder as audio)
--models-dir <dir>	Model cache directory (default: ./models)
--mode streaming|offline	Diarization mode (default: offline)
--threshold <float>	Speaker clustering threshold (default: 0.6)
--model-version v2|v3	ASR model — default v3 (multilingual); use v2 for English-only
--model-dir <path>	Use local ASR models instead of downloading
--speakers <n1,n2,...>	Initial speaker display names
-h, --help	Show help

Output

Given meeting.wav, MecoScribe produces:

• meeting.txt — readable transcript:

  [00:12] Speaker 1:
  Welcome everyone to today's meeting.
  
  [00:18] Speaker 2:
  Thanks for having me.
  

• meeting.html — open in any browser. The HTML references the original audio file via a relative path, so keep both files together (or open the HTML from the same directory).

How it works

1. Diarization — FluidAudio identifies who spoke when (offline VBx pipeline by default)
2. Transcription — Parakeet ASR with word-level timestamps
3. Alignment — words are mapped to speakers by timestamp overlap
4. Export — plain text and self-contained HTML are written

License

MecoScribe is licensed under the MIT License.

FluidAudio models and runtime are subject to their respective licenses (MIT / Apache 2.0). See the FluidAudio repository for details.