IndicSUPERB

by AI4Bharat (IIT Madras)

Indic-language version of SUPERB — 12 languages × 6 speech tasks.

TL;DR

Indic-language version of SUPERB — 12 languages × 6 speech tasks.

Best for probing Indic SSL speech models across LID / ASR / SV / IC / KS / QbE. Pricing: free.

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

What it is

IndicSUPERB extends SUPERB to 12 Indian languages across 6 tasks. The standard Indic SSL probe. License: Apache-2.0 toolkit.

Best for: Probing Indic SSL speech models across LID / ASR / SV / IC / KS / QbE.
Watch out for: Apache-2.0 toolkit · component data is Kathbath + Common Voice + others. Cite: Javed et al., AAAI 2023.

Install / use

git clone https://github.com/AI4Bharat/IndicSUPERB

Features

Speaker diarizationNo
Word-level timestampsNo
Streaming / real-timeNo
Languages supported12
HIPAA eligibleNo

IndicSUPERB vs Whipscribe

FeatureIndicSUPERBWhipscribe
CategoryOpen sourceTranscription APIs
Pricingfreefree beta
Speaker diarizationNoYes
Word timestampsNoYes
StreamingNoNo
Languages1299
PlatformsGitHubWeb, API, MCP

Alternatives to IndicSUPERB

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.