> ## Documentation Index
> Fetch the complete documentation index at: https://docs.babbl-labs.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Overview

The Extended Dataset provides granular transcript-level data from YouTube financial media, offering near complete segment-by-segment coverage (approximately 85%) with detailed speaker information and comprehensive metadata. This dataset is ideal for deep linguistic analysis, speaker tracking, and fine-grained content analysis across YouTube's financial media landscape.

## Complete Data Dictionary

| Field                                | Type      | Description                                                          |
| ------------------------------------ | --------- | -------------------------------------------------------------------- |
| **segment\_id**                      | UUID      | Primary Key - Globally unique identifier for each transcript segment |
| **video\_id**                        | UUID      | Unique YouTube video identifier                                      |
| **channel\_id**                      | UUID      | Immutable Babbl Labs internal unique identifier for channel          |
| **channel\_uri**                     | UUID      | Immutable YouTube's unique identifier for channel                    |
| **channel\_custom\_url**             | STRING    | Mutable custom URL for channel                                       |
| **channel\_name**                    | STRING    | Channel title from YouTube metadata                                  |
| **channel\_description**             | STRING    | Channel description from YouTube metadata                            |
| **channel\_locale**                  | ENUM      | Channel geographic country (ISO 3166-1 alpha-2)                      |
| **channel\_published\_at**           | TIMESTAMP | Timestamp when channel was created                                   |
| **channel\_coverage\_initiated\_at** | TIMESTAMP | Timestamp when we initiated coverage                                 |
| **video\_title**                     | STRING    | Video title from YouTube metadata                                    |
| **video\_description**               | STRING    | Video description from YouTube metadata                              |
| **video\_language**                  | STRING    | Video language code (ISO 639-1)                                      |
| **video\_published\_dt**             | TIMESTAMP | Timestamp when video was originally published                        |
| **video\_download\_dt**              | TIMESTAMP | Timestamp when we downloaded the video                               |
| **video\_transcribed\_at**           | TIMESTAMP | Timestamp when we transcribed the video                              |
| **video\_in\_dataset\_at**           | TIMESTAMP | Timestamp when video was included in dataset                         |
| **model\_transcription\_tag**        | STRING    | Identifier for transcription model used                              |
| **segment\_start**                   | FLOAT     | Starting point of segment in seconds from video start                |
| **segment\_end**                     | FLOAT     | End point of segment in seconds from video start                     |
| **segment\_start\_char**             | INT       | Character index where segment starts in transcript                   |
| **segment\_end\_char**               | INT       | Character index where segment ends in transcript                     |
| **segment\_text**                    | STRING    | Complete verbatim transcript text for this segment                   |
| **speaker\_name**                    | STRING    | Name of speaker (optional - may be null)                             |