openai-whisper

To transcribe audio files into text or subtitles, this Skill converts spoken content from meetings or podcasts into plain text, SRT, VTT, or JSON formats using local or cloud processing.

Yes, you can run speech-to-text transcription locally on-device for privacy-sensitive audio without needing an API key, ensuring secure processing of meetings and interviews.

Yes, it supports translating foreign-language audio into English text. Using the OpenAI cloud API, you can configure language hints to generate translated transcripts and subtitles.

The best way to generate timestamped subtitles for a published video is to transcribe your recorded audio and export the results directly as SRT or VTT files for downstream publishing.

Available output formats for structured transcript data include plain text, SRT, VTT, JSON, and TSV, supporting downstream editing, publishing, and analysis of transcribed meetings.

Use cloud transcription instead of local on-device transcription when speed matters for faster processing or batch translation tasks, whereas local transcription suits privacy-sensitive audio.