Fairseq

by Meta AI

Meta's seq-to-seq toolkit — home of wav2vec, HuBERT, XLS-R, MMS.

TL;DR

Meta's seq-to-seq toolkit — home of wav2vec, HuBERT, XLS-R, MMS.

Best for researchers needing the original implementations of wav2vec / HuBERT / MMS. Pricing: free.

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

What it is

The canonical research codebase for many SOTA speech encoders. MIT.

Best for: Researchers needing the original implementations of wav2vec / HuBERT / MMS.
Watch out for: Maintenance-mode in favor of fairseq2; old PyTorch versions; setup pain.

Install / use

pip install fairseq

Features

Speaker diarizationNo
Word-level timestampsYes
Streaming / real-timeNo
Languages supported100
HIPAA eligibleNo

Fairseq vs Whipscribe

FeatureFairseqWhipscribe
CategoryOpen sourceTranscription APIs
Pricingfreefree beta
Speaker diarizationNoYes
Word timestampsYesYes
StreamingNoNo
Languages10099
PlatformsLinuxWeb, API, MCP

Alternatives to Fairseq

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.