faster-whisper

Transcribe audio files locally with faster-whisper and CTranslate2.

9|2|Updated Jan 28, 2026

Search Tags:#transcription #whisper #audio #gpu #speech-to-text #cuda

Ready to run

npx skills add https://github.com/ThePlasmak/faster-whisper --skill faster-whisper-theplasmak

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: faster-whisper
Source: https://github.com/ThePlasmak/faster-whisper/tree/main
Command: npx skills add https://github.com/ThePlasmak/faster-whisper --skill faster-whisper-theplasmak

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides a highly efficient solution for converting spoken audio into written text, significantly faster than traditional methods.

Core Features & Use Cases

Rapid Transcription: Transcribe audio files in a fraction of the time, especially with GPU acceleration.
High Accuracy: Achieves accuracy comparable to OpenAI's Whisper model.
Versatile: Supports various audio formats, multilingual transcription, and provides word-level timestamps.
Use Case: Quickly generate transcripts for meeting recordings, interviews, or podcasts, enabling faster content creation and analysis.

Quick Start

Use the faster-whisper skill to transcribe the attached audio file 'meeting.mp3'.

Frequently Asked Questions about faster-whisper

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I transcribe audio files locally with GPU acceleration?▼

To transcribe audio files locally with GPU acceleration, use the faster-whisper library, which leverages CTranslate2 to deliver 4-6x faster processing speeds while maintaining identical accuracy to standard Whisper models.

What is the best way to speed up speech-to-text transcription?▼

Speeding up speech-to-text transcription is achieved by using faster-whisper, a CTranslate2 reimplementation that runs significantly faster than traditional methods by supporting GPU acceleration for near real-time performance.

Can I get word-level timestamps from audio transcription?▼

Yes, word-level timestamps are supported by this transcription approach, which also includes multilingual content handling and voice activity detection for silence removal to provide accurate and structured text output.

Does faster-whisper support multilingual audio transcription?▼

Multilingual audio transcription is supported by faster-whisper, allowing you to process various audio formats and convert spoken content into text with high accuracy comparable to OpenAI's Whisper model.

Do I need a GPU to run faster-whisper for speech-to-text?▼

A GPU is not strictly required to run faster-whisper for speech-to-text, but using GPU acceleration significantly increases processing speeds, enabling near real-time performance for rapid transcription of meetings or podcasts.