ScrAIbe – LocalAI-Backed Transcription and Summarization

ScrAIbe is a transcription and summarization service that:

• Sends audio to a LocalAI server running vibevoice.cpp for transcription and speaker diarization.
• Optionally uses a second LLM to generate a structured summary.
• Provides:
  • A web GUI for uploading audio and receiving transcripts via email.
  • A CLI and Python API for direct integration.

No local speech models or heavy dependencies are required. ScrAIbe is designed as a thin client in front of your own AI services.

For more information: https://apstrom.ca

Features

• Transcription with speaker diarization via LocalAI:
  • Uses the /v1/audio/diarization endpoint.
  • Compatible with vibevoice.cpp and other diarization-capable backends.
• Optional AI-powered summarization:
  • Task: transcript_and_summarize
  • Highlights:
    • Main topics and discussion points
    • Key decisions and outcomes
    • Action items and responsibilities
    • Open issues and risks
• Async web GUI:
  • Upload audio via browser.
  • Jobs are queued and processed in the background (Celery + Redis).
  • Emails:
    • Immediate confirmation with queue position.
    • Final transcript (MD + DOCX + JSON) when ready.
    • Summary as MD + DOCX (if requested).
    • Error notification if processing fails.
• File formats:
  • Transcript: .md and .docx (line-numbered, no cover page)
  • Summary (if requested): .md and .docx (no line numbering, no cover page)
  • Full structured output: .json
• Customizable branding:
  • Web GUI title, logo, and accent color via environment variables.
  • Email logo, accent color, and subject lines via environment variables.
• CLI and Python API:
  • Simple command-line interface.
  • Drop-in Scraibe class for integration into other tools.
• Docker-ready:
  • Lightweight container, configured via environment variables.

Architecture

• LocalAI (vibevoice.cpp):
  • Handles audio → transcript + speaker segments.
• Summarizer LLM (OpenAI-compatible chat endpoint):
  • Handles transcript → structured summary.
• ScrAIbe:
  • Orchestrates:
    • File upload to LocalAI
    • Transcript assembly
    • Chunked summarization
    • Output formatting (e.g., .md with transcript + summary)
  • Runs:
    • Web GUI (Gradio)
    • Celery worker (async processing)
    • Redis (in-container by default)

Quick Start (Web GUI in Docker)

Run the container with your LocalAI and summarizer endpoints:

• docker run -d
  -p 7860:7860
  -e LOCALAI_API_URL=http://localai:8080
  -e SUMMARIZER_API_URL=http://llm:8080
  -e EMAIL_SMTP_HOST=smtp.your-domain.com
  -e EMAIL_SMTP_PORT=587
  -e EMAIL_SMTP_USER=transcribe@your-domain.com
  -e EMAIL_SMTP_PASSWORD=your_password
  -e EMAIL_FROM_ADDRESS="ScrAIbe transcribe@your-domain.com"
  -e EMAIL_CONTACT_ADDRESS=support@your-domain.com
  -e WEBUI_TITLE="Your Transcription Service"
  -e WEBUI_LOGO_URL="https://your-domain.com/logo.png"
  -e EMAIL_LOGO_URL="https://your-domain.com/logo.png"
  -e EMAIL_ACCENT_COLOR="#7C6DA0"
  scraibe:latest

Then open: http://:7860

Quick Start (CLI)

Basic usage:

• Transcribe:

  • python3 -m scraibe.cli -f "audio.wav" -o "./output" -of txt

• Transcribe and summarize:

  • python3 -m scraibe.cli -f "audio.wav" -o "./output" --task transcript_and_summarize

Environment variables must be set to point to your LocalAI and summarizer LLM.

Python API

Example: transcribe only

• from scraibe import Scraibe

  • client = Scraibe()
  • text = client.transcribe("audio.wav")
  • print(text)

Example: transcribe and summarize

• from scraibe import Scraibe

  • client = Scraibe()
  • result = client.transcript_and_summarize("audio.wav")
  • transcript = result["transcript"]
  • summary = result["summary"]

You can override endpoints and models via environment variables or constructor parameters if needed.

Command-Line Options

Run:

• python3 -m scraibe.cli -h

Key options:

• -f / --audio-files:
  • One or more audio files to process.
• --task:
  • transcribe (default)
  • transcript_and_summarize
• -o / --output-directory:
  • Output folder for generated files.
• -of / --output-format:
  • txt, json, md, html
  • For transcript_and_summarize, output is always saved as .md with:

    • Transcript

    • Summary

Other options (e.g., --language, --num-speakers) are accepted and forwarded where applicable; many legacy Whisper/Pyannote flags are kept for compatibility but ignored.

Docker Usage

ScrAIbe is designed to run in Docker as a client to your LocalAI and summarizer LLM.

Basic run (transcribe via CLI)

• docker run -it
  -e LOCALAI_API_URL=http://localai:8080
  -v /path/to/audio:/audio
  scraibe:latest
  -f /audio/meeting.wav -o /audio/output -of txt

Basic run (transcribe + summarize via CLI)

• docker run -it
  -e LOCALAI_API_URL=http://localai:8080
  -e SUMMARIZER_API_URL=http://llm:8080
  -v /path/to/audio:/audio
  scraibe:latest
  -f /audio/meeting.wav -o /audio/output --task transcript_and_summarize

Docker Environment Variables

The following environment variables configure ScrAIbe in Docker.

Transcription / Diarization (LocalAI):

• LOCALAI_API_URL:
  • Required.
  • Base URL of the LocalAI server.
  • Example: http://localai:8080
• LOCALAI_API_KEY:
  • Optional.
  • API key for LocalAI, if configured.
• LOCALAI_MODEL:
  • Optional (default: vibevoice-diarize).
  • Model name used for transcription/diarization.

Summarization LLM:

• SUMMARIZER_API_URL:
  • Required when using --task transcript_and_summarize.
  • Base URL of the summarization LLM (OpenAI-compatible /v1/chat/completions).
  • Example: http://llm:8080
• SUMMARIZER_API_KEY:
  • Optional.
  • API key for the summarization LLM, if required.
• SUMMARIZER_MODEL:
  • Optional (default: llama-3.1-8b-instruct).
  • Model name used for summarization.

Web GUI and branding:

• WEBUI_TITLE:
  • Title shown in the web GUI (default: A.P.Strom Transcription).
• WEBUI_LOGO_URL:
  • URL of the logo displayed in the web GUI header.
  • Example: https://your-domain.com/logo.png

Accent color (UI and emails):

• EMAIL_ACCENT_COLOR:
  • Accent color used in:
    • Web GUI buttons and accents
    • Email headings, links, and email addresses
  • Default: #7C6DA0

Async processing (Celery + Redis):

• CELERY_BROKER_URL:
  • Redis broker URL (default: redis://localhost:6379/0).
• CELERY_RESULT_BACKEND:
  • Redis backend URL (default: redis://localhost:6379/0).
• SCRAIBE_UPLOAD_DIR:
  • Directory where uploaded audio is stored (default: /tmp/scraibe_uploads).

Email configuration:

• EMAIL_SMTP_HOST:
  • SMTP server host.
• EMAIL_SMTP_PORT:
  • SMTP server port (e.g., 587).
• EMAIL_SMTP_USER:
  • SMTP username.
• EMAIL_SMTP_PASSWORD:
  • SMTP password.
• EMAIL_SMTP_USE_TLS:
  • Use TLS (true/false; default: true).
• EMAIL_FROM_ADDRESS:
  • Sender address (e.g., "ScrAIbe transcribe@your-domain.com").
• EMAIL_CONTACT_ADDRESS:
  • Support contact address shown in email templates.
• EMAIL_LOGO_URL:
  • URL of the logo used in emails (preferred).
• EMAIL_LOGO_PATH:
  • Fallback local path for email logo (default: /app/src/misc/logo1.png).
• EMAIL_CSS_PATH:
  • Path to the CSS used in emails (default: /app/src/misc/mail_style.css).

Email subject lines (customizable):

• EMAIL_SUBJECT_UPLOAD:
  • Subject for upload confirmation email.
  • Default: "ScrAIbe: Your transcription request has been received"
• EMAIL_SUBJECT_SUCCESS:
  • Subject for transcript-ready email.
  • Default: "ScrAIbe: Your transcript is ready"
• EMAIL_SUBJECT_ERROR:
  • Subject for error notification email.
  • Default: "ScrAIbe: Error with your transcription request"

Output files (async web GUI):

When a job completes, the user receives:

• Transcript:
  • .md file
  • .docx file (line-numbered, no cover page)
• Summary (if requested):
  • .md file
  • .docx file (no line numbering, no cover page)
• JSON:
  • Structured transcript with diarization and metadata

All of these can also be overridden from the CLI when needed (e.g., --localai-api-url, --summarizer-api-url).

Dependencies

Core runtime dependencies:

• Python 3.9+
• httpx
• numpy
• tqdm
• gradio
• celery[redis]
• redis
• python-docx
• ffmpeg (for audio preprocessing)

No local Whisper, PyTorch, or Pyannote models are required.

Contributing

Contributions are welcome. Please refer to CONTRIBUTING.md for guidelines.

License

This project is licensed under GPL-3.0. See LICENSE for details.