whisper

Transcribe and translate audio across 99 languages using Whisper.

Updated Apr 3, 2026

Search Tags:#translation #transcription #whisper #multilingual #audio-processing #speech-recognition #model-sizes

Ready to run

npx skills add https://github.com/handsomelong922/my-codex-skills --skill whisper-handsomelong922

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: whisper
Source: https://github.com/handsomelong922/my-codex-skills/tree/main/skills/whisper
Command: npx skills add https://github.com/handsomelong922/my-codex-skills --skill whisper-handsomelong922

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires openai-whisper, transformers, torch, and includes references (resource) components.

What problem does it solve?

Whisper enables automatic transcription and translation of long-form audio across 99 languages, turning speech into searchable, editable text for diverse workflows.

Core Features & Use Cases

Multilingual ASR across 99 languages with scalable model sizes from tiny to large for speed and accuracy.
On-demand translation to English and language identification to support cross-language content pipelines.
Use cases include podcast transcription, meeting notes, video captioning, and archival audio indexing.

Quick Start

Install the Whisper package and run a transcription on an audio file to generate text.

Frequently Asked Questions about whisper

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

FAQPage Schema

How do I transcribe podcast audio to text using multilingual speech recognition?▼

Whisper transcribes podcast audio to text across 99 languages using scalable model sizes from tiny to large. It requires installing openai-whisper, transformers, and torch, with optional GPU acceleration to speed up processing long-form audio recordings.

Can I translate spoken audio to English text from other languages?▼

Yes, Whisper translates spoken audio to English text from 99 languages. Beyond multilingual transcription, it performs on-demand translation to English and language identification to support cross-language content pipelines for meetings, podcasts, and video captioning.

Do I need a GPU for fast audio processing and speech-to-text transcription?▼

GPU acceleration is optional for audio processing and speech-to-text transcription with Whisper, but significantly improves speed. CPU processing works, but running large model sizes for higher accuracy on long-form audio benefits greatly from GPU acceleration.

What is the best way to generate video captioning from audio files?▼

The best way to generate video captioning from audio files is using Whisper for multilingual speech-to-text transcription. It supports 99 languages and scalable model sizes, enabling automated generation of searchable, editable text from spoken content for video captioning workflows.

Does Whisper support automated meeting notes transcription for different languages?▼

Yes, Whisper supports automated meeting notes transcription across 99 languages. It applies multilingual ASR to turn spoken meeting audio into searchable, editable text, with model sizes ranging from tiny to large to balance speed and accuracy for diverse workflows.

When should I use smaller model sizes instead of large models for transcription?▼

Use smaller Whisper model sizes when transcription speed is critical or resources are limited, and large models when maximum accuracy is needed. Scalable model sizes from tiny to large let you balance processing speed and transcription accuracy across 99 languages.