wq2012/SimpleDER
A lightweight library to compute Diarization Error Rate (DER).
Supports overlapped speech detection and uses the Hungarian algorithm for optimal speaker-to-speaker assignment between reference and hypothesis annotations. The implementation applies optional collar-based boundary smoothing and computes DER via a mathematically equivalent "Load-Match" formulation that accounts for varying numbers of simultaneous speakers. Integrates with `scipy.optimize.linear_sum_assignment` and provides a minimal Python API accepting speaker diarization segments as tuples of (speaker_id, start_time, end_time).
Available on PyPI.
Stars
62
Forks
9
Language
Python
License
Apache-2.0
Category
Last pushed
Jan 14, 2026
Monthly downloads
611
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/wq2012/SimpleDER"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related tools
fgnt/meeteval
MeetEval - A meeting transcription evaluation toolkit
readbeyond/aeneas
aeneas is a Python/C library and a set of tools to automagically synchronize audio and text (aka...
analyticsinmotion/werpy
🐍📦 Ultra-fast Python package for calculating and analyzing the Word Error Rate (WER). Built for...
kahne/fastwer
A PyPI package for fast word/character error rate (WER/CER) calculation
tabahi/bournemouth-forced-aligner
Extract phoneme-level timestamps from speeh audio.