index-deposition

Name: index-deposition
Availability: InStock
Author: jdrodriguez

Transcribe deposition recordings and generate topic indexes with key moments.

6|Updated Feb 10, 2026

Search Tags:#timeline #transcription #litigation #topic-indexing #deposition #speaker-diarization #key-moments

Ready to run

npx skills add https://github.com/jdrodriguez/legal-toolkit --skill index-deposition

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: index-deposition
Source: https://github.com/jdrodriguez/legal-toolkit/tree/main/legal-toolkit/skills/index-deposition
Command: npx skills add https://github.com/jdrodriguez/legal-toolkit --skill index-deposition

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires faster-whisper, pydub, plotly, pandas, XlsxWriter, python-dateutil, torch, and includes scripts (resource) components.

What problem does it solve?

Depositions often generate lengthy, unstructured media and transcripts that are hard to search or review. This skill transcribes deposition recordings, identifies speakers, builds topic indexes, and flags key moments to support efficient review and litigation preparation.

Core Features & Use Cases

Transcribes deposition recordings with timestamps and speaker labels
Builds topic indexes from testimony to enable quick topic search
Detects and highlights key moments such as admissions, objections, and contradictions
Outputs include transcript, page:line transcript, topic_index.json, key_moments.json, and an interactive testimony timeline

Quick Start

Provide the path to a deposition recording (audio or video) to index and generate outputs.

Frequently Asked Questions about index-deposition

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

FAQPage Schema

How do I transcribe and index a deposition recording with timestamps?▼

Deposition indexing transcribes audio or video recordings using faster-whisper, adds timestamps and speaker labels, and outputs searchable transcripts, topic indexes, and key moments. Provide the path to your deposition recording to generate structured litigation support files.

Can I generate page:line transcripts from deposition video files?▼

Yes, deposition video files can be processed to generate page:line transcripts. The Skill transcribes the media, organizes the text with timestamps, and outputs a formatted transcript alongside a topic_index.json and key_moments.json for litigation preparation.

Does faster-whisper support speaker diarization for litigation transcripts?▼

Faster-whisper handles the transcription, while optional pyannote.audio provides speaker diarization to identify speakers. Combined with pydub for audio processing, this produces indexed transcripts with speaker labels for litigation review.

What is the best way to flag admissions and objections in deposition audio?▼

Indexing deposition recordings automatically detects and highlights key moments such as admissions, objections, and contradictions. It outputs a key_moments.json file and an interactive testimony timeline visualization to support efficient review.

What file formats are output when indexing deposition recordings?▼

Indexing deposition recordings outputs a full transcript, a page:line transcript, topic_index.json, key_moments.json, and an interactive testimony timeline visualization. These outputs enable quick topic search and testimony review for litigation support.

Do I need pyannote.audio to transcribe depositions, or is faster-whisper enough?▼

Faster-whisper is required for transcription, while pyannote.audio is optional for speaker diarization. Faster-whisper alone transcribes and timestamps the deposition, but adding diarization identifies speakers to produce labeled litigation transcripts.