Skip to content
TakoAPI
Back to skills

iyeque-audio-processing

by iyeque

AI & LLMs
—Not yet rated
Your rating

Brief

Audio ingestion, analysis, transformation, and generation (Transcribe, TTS, VAD, Features).

Description

What This Skill Does

A toolset for audio ingestion, analysis, transformation, and generation. Supports speech-to-text via OpenAI Whisper, text-to-speech via Google TTS, feature extraction, voice activity detection, and audio transformations like trimming and resampling. File paths are validated and system directories are blocked.

Combines transcription, synthesis, VAD, and transformation in one unified tool instead of wiring together separate CLI tools manually.

When to Use It

  • Transcribe a recorded meeting or interview to text
  • Generate spoken audio from a script or document
  • Find speech segments in a long audio file
  • Extract acoustic features for audio classification
  • Trim and normalize audio clips for editing pipelines

Example Workflow

INPUT

User asks: Transcribe this podcast episode and find where people are speaking

AGENT
  1. 1

    Run vad_segments on the audio file to detect speech segments

  2. 2

    Review segment timestamps to identify active speech regions

  3. 3

    Run transcribe with the base Whisper model on the full file

  4. 4

    Return transcript text along with segment-level timestamps

OUTPUT

Full transcript with time-coded speech segments

Installation

clawhub install iyeque/iyeque-audio-processing

Run the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.

Links