whisper

Transcribe and translate audio across 99 languages with openai-whisper.

Updated Apr 19, 2026
One-click install
npx skills add https://github.com/gqf2008/hermez-ai --skill whisper-gqf2008
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: whisper
Source: https://github.com/gqf2008/hermez-ai/tree/main/skills/mlops/models/whisper
Command: npx skills add https://github.com/gqf2008/hermez-ai --skill whisper-gqf2008

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Whisper automates the conversion of spoken language into written text across 99 languages, enabling transcription, translation to English, and language identification in a single model.

Core Features & Use Cases

  • Multilingual transcription: Transcribe speech to text in 99 languages.
  • Translation to English: Translate transcripts into English for broader accessibility.
  • Language identification: Detect the language of the input audio.
  • Flexible model sizes: Six model sizes from tiny to large for a trade-off between speed and accuracy.
  • Use cases: Podcast transcription, meeting notes automation, and multilingual audio processing in media workflows.

Quick Start

Install the Whisper package and run a sample transcription on your audio file.

Frequently Asked Questions about whisper

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

FAQPage Schema
How do I transcribe audio to text across multiple languages using a single model?

Multilingual transcription is handled by a single model covering 99 languages, converting spoken audio into written text. It also performs language identification to detect the input audio's language automatically.

Can I translate non-English audio files into English text?

Yes, translation to English is supported directly. The model transcribes spoken audio from any of the 99 supported languages and outputs the resulting text translated into English for broader accessibility.

Do I need openai-whisper, transformers, and torch installed for speech-to-text processing?

Yes, this speech-to-text processing requires openai-whisper, transformers, and torch. These dependencies provide the underlying deep learning framework and model architecture needed to run the audio transcription workflows.

What's the best way to balance transcription speed and accuracy for podcast audio?

Balance speed and accuracy by choosing from six model sizes, tiny to large. Smaller models process faster for quick podcast transcription, while larger models deliver higher accuracy for complex audio workflows.

Does this speech-to-text approach work for meeting notes automation?

Yes, meeting notes automation is a supported use case. The model transcribes spoken audio into written text, making it applicable for automatically generating text records from recorded meetings and multilingual audio processing.