skills.rest
yousufjoyian
yousufjoyianCOMMUNITY
Explore all skills by this author →
GitHub

audio-transcriber

Transcribe and diarize dashcam video audio into timestamped transcripts and speaker labels.

Updated Aug 27, 2026
Search Tags:#transcription#JSON#GPU#diarization#SRT#INDEX.csv#dashcam
One-click install
npx skills add https://github.com/yousufjoyian/claude-skills --skill audio-transcriber
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: audio-transcriber
Source: https://github.com/yousufjoyian/claude-skills/tree/main/Dashcam
Command: npx skills add https://github.com/yousufjoyian/claude-skills --skill audio-transcriber

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires ffmpeg-python, torch, torchaudio, faster-whisper, pyannote.audio, speechbrain, psutil, cpuinfo, librosa, soundfile, pynvml, numpy, pydub, webvtt-py, pysrt, yaml, python-dotenv, click, rich, pandas, scipy, faiss-cpu, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill converts dashcam video audio into accurate, timestamped transcripts and optionally assigns speaker labels, enabling quick review and searchable archives of driving conversations.

Core Features & Use Cases

  • GPU-accelerated transcription: Uses faster-whisper with CUDA to transcribe large dashcam video collections quickly.
  • Speaker diarization (optional): Produces unified speaker labels (spkA, spkB, etc.) across transcripts for multi-speaker sessions.
  • Multi-format outputs: Generates TXT, JSON, SRT, and optional VTT transcripts, plus a global INDEX.csv for search across days.
  • Date/folder organization: Outputs organized by date and video, with a central results JSON and per-day indices for cross-day analysis.
  • Use Case: Imagine a fleet manager needing to review driver conversations across hundreds of drives; this Skill batches the work, tags who spoke when, and provides easily searchable transcripts.

Quick Start

  1. Place your dashcam videos in a date-stamped directory (e.g., MYY/MM/DD or YYYYMMDD within a base folder).
  2. Run the skill to transcribe a folder, enable optional diarization, and export TXT/JSON/SRT (and INDEX.csv) to a chosen outputs directory.
  3. Inspect the produced INDEX.csv to search phrases across all transcripts and jump to exact time stamps.

Frequently Asked Questions about audio-transcriber

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

FAQPage Schema
How do I transcribe dashcam video audio with GPU acceleration?▼

GPU-accelerated transcription uses faster-whisper with CUDA to convert dashcam video audio into timestamped text transcripts quickly. This Skill automates the full pipeline: it discovers videos, extracts audio, segments it, transcribes with faster-whisper, and outputs TXT, JSON, and SRT formats for easy review and searching across multiple days of footage.

Can I add speaker labels to dashcam transcripts?▼

Yes, optional speaker diarization assigns unified speaker labels (spkA, spkB, etc.) across transcripts using pyannote.audio or speechbrain backends. This identifies who spoke when, enabling fleet managers to track driver conversations and attribute statements to specific speakers in multi-speaker sessions.

How do I search across multiple days of dashcam transcripts?▼

The Skill generates a centralized INDEX.csv and global results JSON that indexes all transcripts from your date-organized video folders. Search this index by phrase to find exact timestamps across hundreds of drives, enabling quick cross-day analysis without reviewing individual files.

What are the GPU and system requirements for this transcription pipeline?▼

You need a CUDA-enabled GPU, ffmpeg, Python 3.x, and dependencies including torch, torchaudio, faster-whisper, and optional diarization libraries (pyannote.audio or speechbrain). The Skill monitors GPU metrics with pynvml and CPU usage with psutil to ensure safe, resumable runs on your hardware.

What output formats does the transcription produce?▼

The Skill generates TXT, JSON, SRT, and optional WebVTT transcripts per video, plus a global INDEX.csv for cross-day search. Date-organized outputs include per-day indices and a centralized results JSON, enabling flexible export for downstream review tools and fleet management systems.

Does this work for autonomous driving and fleet workflows?▼

Yes, the Skill targets autonomous driving and fleet review use cases. It batches large dashcam collections, tags speakers, and produces searchable archives that let fleet managers review driver conversations across hundreds of drives without manual transcription overhead.

Related Skills

Explore More Data & Analytics Skills→
cici

transcribe

Transcribes local audio files to text using NVIDIA Parakeet (NeMo).

Community
Advanced
hmzainjamilhmzainjamil
1

Transcription Automation

Convert audio and video recordings into transcripts, subtitles, and meeting notes.

Community
Advanced
secondskysecondsky
204

ml-model-training

Train machine learning models with scikit-learn, PyTorch, and TensorFlow.

Community
Advanced
akrindevakrindev

video-transcriber

Transcribe video audio into timestamped SRT, VTT, or JSON transcripts.

Community
Advanced