SPGISpeech

by Kensho Technologies (S&P Global)

5000h of professionally-transcribed earnings-call audio — financial-domain ASR.

TL;DR

5000h of professionally-transcribed earnings-call audio — financial-domain ASR.

Best for domain-adapted ASR for financial/earnings-call transcription; non-native English accents. Pricing: research-only.

Category
Open source
License
—
Stars
—
Last push
—
Pricing
research-only
Platforms
Web

What it is

SPGISpeech is 5000 hours of professional-grade earnings-call transcription from Kensho. Earnings calls feature heavy non-native accents and finance jargon — the canonical financial-domain ASR benchmark. Research-only license.

Best for: Domain-adapted ASR for financial/earnings-call transcription; non-native English accents.
Watch out for: Custom license · NON-COMMERCIAL ONLY (Kensho research license) · registration + agreement required · ~50k speakers across ~16 industries. Cite: O'Neill et al., Interspeech 2021.

Install / use

https://datasets.kensho.com/datasets/spgispeech  # registration required

Features

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

SPGISpeech vs Whipscribe

FeatureSPGISpeechWhipscribe
CategoryOpen sourceTranscription APIs
Pricingresearch-onlyfree beta
Speaker diarizationNoYes
Word timestampsNoYes
StreamingNoNo
Languages199
PlatformsWebWeb, API, MCP

Alternatives to SPGISpeech

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.