Audio · Video / Public Sector & Legal
Labs YouTube Transcription Database | YouTube Video Transcript Firehose | EN/US | 25K+ Channels, Sentiment, Speakers, Entities
Advertised by providerLicensing rights not independently verifiedPaid licence
Overview
Labs YouTube Transcription Database | YouTube Video Transcript Firehose | EN/US | 25K+ Channels, Sentiment, Speakers, Entities is an audio and video dataset. Typical applications include speech recognition, speaker recognition, LLM pre-training and 2 more. Relevant sectors: Public Sector & Legal, Media & Entertainment, Consumer & Lifestyle and Financial Services. The listing mentions transcription and speaker labels. It is offered as a paid dataset; pricing details are provided on request.
Primary applications
What the listing offers
- Listing mentions transcription, speaker labels.
TranscriptDB is the full YouTube transcript firehose, built for systematic hedge funds with in-house NLP capacity and AI labs training foundation models. Smaller buyers should look at Tripwire instead. What you get: - 25,000+ pre-filtered market-relevant YouTube channels - 1M+ videos processed monthly - Intraday delivery (2-3 hour clip latency) - Named entities, speaker diarization, sentiment scoring on every transcript - 5+ years of historical archive available as a separate backfile Use cases: - Quantitative alpha generation and signal research - Foundation model training and fine-tuning corpora - Long-form NLP backtesting and event studies - Cross-source entity disambiguation at scale
Related datasets
More in Video & Physical AI →AudioAdvertised
Multilingual Podcast Audio Dataset - 57K Hours, 12 Languages | Audio Data
This dataset includes a large-scale collection of podcast audio data across multiple languages, domains, and speaker demographics.
- Language
- —
- Volume
- 57k h
- Domain
- Speech & Audio
- Use case
- Speech recognition
Price on request
More speech & audio datasets
All 1,727 speech & audio datasets →- Labs YouTube Tripwire | CEO & Executive Commentary Feed from YouTube | US/EN | 5K+ Companies, Intraday Sentiment Data & Speaker Context
- Labs YouTube Tripwire | S&P 500 Ticker-Mapped YouTube Mentions | US/EN | 25K+ Companies, Intraday Sentiment & Entity Tags
- Lao live conversational audio, channel-separated
- Lao live conversational audio, single-channel
- Lao Scripted Monologue Smartphone Speech Dataset | 500 Hours | Multilingual Language Data
- Large-Scale Egyptian Arabic Multi-Turn Conversational Speech Dataset for Daily Communication and Opinion Exchange
- Large-Scale English Business Multi-Turn Conversational Speech Dataset
- Large-Scale Gulf Arabic Multi-Turn Conversational Speech Dataset for Daily Communication and Opinion Exchange