listen

Transcribes speech and analyzes audio locally using whisper and librosa without network calls.

621|58|Updated Mar 15, 2026
One-click install
npx skills add https://github.com/Lingtai-AI/lingtai --skill listen-lingtai-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: listen
Source: https://github.com/Lingtai-AI/lingtai/tree/main/tui/internal/preset/skills/listen
Command: npx skills add https://github.com/Lingtai-AI/lingtai --skill listen-lingtai-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires faster-whisper, librosa, numpy, and includes scripts (resource) components.

What problem does it solve?

Transforms on-device audio input into structured data: transcription for speech and feature extraction for music, all without cloud dependencies.

Core Features & Use Cases

  • Local transcription of voice notes, lectures, and podcasts using faster-whisper.
  • Music analysis with librosa to estimate tempo, key, spectral properties, and energy contours.
  • Use cases include offline note-taking, lyric extraction (with caveats for singing), and objective audio analysis for QA or research.

Quick Start

Transcribe a voice recording with the on-device Whisper workflow or run music analysis on a local audio file.

Frequently Asked Questions about listen

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

FAQPage Schema
How do I transcribe audio locally without sending data to the cloud?

On-device audio transcription processes your local voice notes entirely offline, avoiding network calls. The skill uses faster-whisper to transcribe speech directly on your machine, ensuring data privacy for sensitive recordings.

Does on-device music analysis work for extracting tempo and key?

Yes, on-device music analysis extracts audio features like tempo, key, and energy contours. Using librosa, it processes local audio files to estimate spectral properties and objective metrics for QA or research.

Do I need to install faster-whisper and librosa before using this?

No manual installation is required beforehand, as the skill installs local dependencies like faster-whisper and librosa automatically on first run. This setup ensures the environment is ready for offline transcription and audio analysis.

Can I extract lyrics from songs using local transcription?

Local transcription can extract lyrics from songs, but with caveats for singing. The faster-whisper workflow is optimized for speech, so vocal pitch variations in music may reduce transcription accuracy.

What are the limitations of on-device audio analysis for offline workflows?

On-device audio analysis limitations include potential slower processing compared to cloud services and reduced accuracy for singing voice transcription. It is best suited for offline note-taking and objective audio feature extraction.

Is offline transcription suitable for privacy-sensitive scenarios?

Yes, offline transcription is ideal for privacy-sensitive scenarios because it does not expose data to external services. All voice-note and music analysis processing occurs locally, ensuring confidential audio remains secure.