Switchboard-1

by LDC

260h of conversational US English telephone speech — historical ASR benchmark.

TL;DR

260h of conversational US English telephone speech — historical ASR benchmark.

Best for conversational telephone ASR baselines; HMM-DNN historical reference. Pricing: paid.

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

What it is

Switchboard-1 (LDC97S62) is 260 hours of US English telephone-pair conversations from 1992 — the foundational conversational-ASR benchmark. LDC paid.

Best for: Conversational telephone ASR baselines; HMM-DNN historical reference.
Watch out for: LDC license · paid · 8 kHz telephone audio · 2-channel · ~260h. Cite: Godfrey et al., ICASSP 1992.

Install / use

https://catalog.ldc.upenn.edu/LDC97S62  # LDC membership

Features

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

Switchboard-1 vs Whipscribe

FeatureSwitchboard-1Whipscribe
CategoryOpen sourceTranscription APIs
Pricingpaidfree beta
Speaker diarizationYesYes
Word timestampsNoYes
StreamingNoNo
Languages199
PlatformsLDCWeb, API, MCP

Alternatives to Switchboard-1

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.