Earnings-22

by Rev.com / Rev.ai

125h earnings-call ASR test set with 27-accent speaker coverage.

TL;DR

125h earnings-call ASR test set with 27-accent speaker coverage.

Best for robustness testing for accented English ASR; financial-domain WER measurement. Pricing: free.

Category
Open source
License
—
Stars
—
Last push
—
Pricing
free
Platforms
HuggingFace, GitHub

What it is

Earnings-22 from Rev is a 125-hour earnings-call test set spanning 27 self-reported English accents. Used as the canonical accent-robustness benchmark. License: CC BY-SA 4.0.

Best for: Robustness testing for accented English ASR; financial-domain WER measurement.
Watch out for: CC BY-SA 4.0 · ~125h test-only (no training split) · accent labels are self-reported. Cite: Del Rio et al., Interspeech 2022.

Install / use

from datasets import load_dataset; ds = load_dataset('revdotcom/earnings22')

Features

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

Earnings-22 vs Whipscribe

FeatureEarnings-22Whipscribe
CategoryOpen sourceTranscription APIs
Pricingfreefree beta
Speaker diarizationYesYes
Word timestampsNoYes
StreamingNoNo
Languages199
PlatformsHuggingFace, GitHubWeb, API, MCP

Alternatives to Earnings-22

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.