insanely-fast-whisper

by Vaibhav Srivastav

CLI that transcribes 150 minutes of audio in ~98 seconds on an A100.

Category
Open source
License
Apache-2.0
Stars
★ 12.4k
Last push
2025-10-25
Pricing
free
Platforms
Linux, GPU

What it is

An opinionated CLI wrapper around Hugging Face Transformers + Flash Attention + BetterTransformer. Trades install complexity for throughput: ~150 min of audio in ~98s on an A100. The reference for "how fast can Whisper go on current hardware." Apache-2.0.

Best for: Batch-processing huge backlogs on rented H100/A100 time.
Watch out for: Requires an NVIDIA GPU with enough VRAM for Whisper-large-v3 + Flash Attention; CPU path is not practical.

Install / use

pipx install insanely-fast-whisper

Features

Speaker diarizationYes
Word-level timestampsYes
Streaming / real-timeNo
Languages supported99
HIPAA eligibleNo

Links

GitHub repo ↗

Alternatives

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, we're one click away.

Try Whipscribe →