NIST SRE Series

by NIST

NIST Speaker Recognition Evaluation — the canonical SV/SD benchmark series.

TL;DR

NIST Speaker Recognition Evaluation — the canonical SV/SD benchmark series.

Best for speaker verification / identification / diarization with NIST gold-standard scoring. Pricing: paid.

Category
Open source
License
—
Stars
—
Last push
—
Pricing
paid
Platforms
LDC

What it is

NIST SRE is the longest-running speaker-recognition challenge series (2004–2024+). Evaluation data via LDC; tooling free. The reference baseline every SV/SD paper compares against.

Best for: Speaker verification / identification / diarization with NIST gold-standard scoring.
Watch out for: LDC-distributed evaluation data · paid · 20+ years of editions (SRE'04..SRE'24). Cite: NIST.gov SRE plans per year.

Install / use

https://www.nist.gov/itl/iad/mig/speaker-recognition-evaluations  # LDC data

Features

Speaker diarizationYes
Word-level timestampsNo
Streaming / real-timeNo
Languages supported1
HIPAA eligibleNo

NIST SRE Series vs Whipscribe

FeatureNIST SRE SeriesWhipscribe
CategoryOpen sourceTranscription APIs
Pricingpaidfree beta
Speaker diarizationYesYes
Word timestampsNoYes
StreamingNoNo
Languages199
PlatformsLDCWeb, API, MCP

Alternatives to NIST SRE Series

OpenAI Whisper
OpenAI

The reference open-source multilingual ASR model from OpenAI.

OSS · MIT ★ 98.1k
whisper.cpp
Georgi Gerganov

C/C++ port of Whisper — runs on anything, from a Raspberry Pi to Apple Silicon.

OSS · MIT ★ 48.8k
faster-whisper
SYSTRAN

4× faster than reference Whisper using CTranslate2 — production sweet spot.

OSS · MIT ★ 22.3k

Whipscribe is a managed faster-whisper + whisperX service. If you want transcripts without running infrastructure, paste a URL or drop a file in the form below — you'll have a transcript in seconds.