AISHELL-2

by Beijing Shell Shell Technology

1000h Mandarin read-speech corpus — scaled-up successor.

TL;DR

1000h Mandarin read-speech corpus — scaled-up successor.

Best for production-scale Mandarin ASR with broader vocabulary. Pricing: research-only.

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

What it is

AISHELL-2 is 1000 hours of Mandarin read speech across iOS / Android / studio microphone channels, 1991 speakers. Research-only license.

Best for: Production-scale Mandarin ASR with broader vocabulary.
Watch out for: Research-license · request form · 1991 speakers · iOS/Android/Mic channel variants. Cite: Du et al., 2018.

Install / use

https://www.aishelltech.com/aishell_2  # request form

Features

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

AISHELL-2 vs Whipscribe

FeatureAISHELL-2Whipscribe
CategoryOpen sourceTranscription APIs
Pricingresearch-onlyfree beta
Speaker diarizationNoYes
Word timestampsNoYes
StreamingNoNo
Languages199
PlatformsWebWeb, API, MCP

Alternatives to AISHELL-2

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.