skills.rest
xu1713
xu1713COMMUNITY
Explore all skills by this author →
GitHub

whisper

Transcribe and translate audio files into text across 99 languages using OpenAI Whisper models.

Updated Jun 5, 2026
Search Tags:#translation#speech recognition#transcription#whisper#audio processing#multilingual#asr
One-click install
npx skills add https://github.com/xu1713/openhorse --skill whisper-xu1713
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: whisper
Source: https://github.com/xu1713/openhorse/tree/main/openhorse/openhorse/optional-skills/mlops/whisper
Command: npx skills add https://github.com/xu1713/openhorse --skill whisper-xu1713

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve? Converting spoken audio into accurate text manually is slow and impractical at scale, especially across multiple languages, noisy recordings, or long podcast and meeting recordings. ## Core Features & Use Cases - Multilingual Transcription: Transcribe speech in 99 languages with six model sizes from tiny (39M params) to large (1550M params), plus a fast turbo variant. - Translation to English: Convert audio in any supported language directly into English text using the translate task. - Flexible Output Formats: Generate plain text, SRT subtitles, WebVTT, or JSON with word-level timestamps via CLI or Python API. - Use Case: Transcribe a batch of recorded podcast episodes into timestamped subtitle files, then feed the text into a LangChain RAG pipeline for searchable show notes. ## Quick Start Ask the AI to transcribe an audio file such as 'meeting.mp3' using the Whisper turbo model and output the result as an SRT subtitle 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 with Whisper in Python?▼

Install openai-whisper and ffmpeg, then load a model with whisper.load_model("base") and call model.transcribe("audio.mp3"). The result dictionary contains the full text plus timestamped segments.

Which Whisper model size should I use?▼

Use turbo for the best speed-to-quality balance and base for quick prototyping. The large model (1550M parameters) gives the highest accuracy but needs about 10 GB of VRAM, while tiny runs fastest with lower accuracy.

Can Whisper translate audio into English?▼

Yes, pass task="translate" to the transcribe method to convert audio in any of the 99 supported languages into English text. Translation only outputs English, regardless of the source language.

Does Whisper support real-time transcription?▼

The standard openai-whisper package is not designed for live captioning due to latency. For streaming or faster inference, use faster-whisper, which runs roughly 4x faster with CTranslate2 optimization.

What are the limitations of Whisper speech recognition?▼

Whisper can hallucinate or repeat text, accuracy degrades on audio longer than 30 minutes, and it provides no speaker diarization. Quality also varies with accents, background noise, and low-resource languages.

Related Skills

Explore More Software Engineering Skills→
buda-aibuda-ai
25

audio-extractor

Extract and enhance audio from video files for MP3/WAV/AAC output.

Official
Intermediate
christian-bromannchristian-bromann
3

langchain-knowledge-base

Build a searchable knowledge base from documents using LangChain.

Community
Advanced
RonanCodesRonanCodes

ingest-video

Transcribe local videos into timestamped text and extracted keyframes.

Community
Intermediate
framerslabframerslab

streaming-stt-deepgram

Transcribe live speech via Deepgram's WebSocket API with speaker diarization.

Official
Advanced