rvrb-transcriber

Atomic audio/video transcription. Audio in, text out.

Part of the reverberage ecosystem — composable LEGO pieces for the newsroom.

Install

# OpenAI API (recommended for quick start)
pip install "rvrb-transcriber[openai]"

# Local whisper (no API calls, runs on your machine)
pip install "rvrb-transcriber[local]"

Requires OPENAI_API_KEY env var when using the openai engine.

Use

CLI

rvrb-transcribe interview.mp3
rvrb-transcribe recording.wav --engine local --language es
rvrb-transcribe video.mp4 --format srt --output subtitles.srt

Python

from rvrb_transcriber import transcribe

result = transcribe("interview.mp3")
print(result.text)

for segment in result.segments:
    print(f"[{segment.start:.1f}s] {segment.text}")

# Export formats
print(result.to_srt())  # SubRip subtitles
print(result.to_vtt())  # WebVTT
print(result.model_dump_json(indent=2))  # JSON

Output formats

Format	Flag
Plain text	--format text (default)
SRT subtitles	--format srt
WebVTT	--format vtt
JSON (with segments)	--format json

Engines

Engine	Requires	Best for
openai	pip install rvrb-transcriber[openai] + API key	Quick, high accuracy, no local GPU
local	pip install rvrb-transcriber[local]	Offline, no API costs, GPU recommended

License

Apache-2.0 — same as the reverberage ecosystem.