Dataset ID:
USE_UHV001
Dataset Name:
English (United States) Ultra High-Volume labeled speech
Common Use Cases:
ASR, Conversational AI, Speech Analytics, Automatic Captioning, In Car HMI & Entertainment, Virtual Assistant
Language:
English
Country:
United States
Language Code:
eng
Country Code:
USA
Product Type
Audio
Detailed Product Type
Broadcast Speech
Unit
1196 hours
Recording Device
N/A
Recording Condition
Low background noise
Contributors
20472
Utterances
423371
Unique Words
110265
Sample Rate (kHz):
16
Channels
1
Data Format
wav
Source
Appen Global
Additional Info:
- Customised packaging available
- High quality labelled speech datasets of web-sourced licensable broadcast audio data, curated to ensure representative speaker demographic distributions, and filtered through human quality checks.
- 12.6M total words
- Utterance-level labelling includes: speech transcription, accent identification, speaker identification, verification, gender and age-group detection, domain classification.
- Domains include: Agriculture & plants, Animals & Pets, Art & Culture, Beauty & Fashion, Career, Clothing, Education, Entertainment, Family & Relationships, Finance & Insurance, Food, Health, History, Hospitality, Legal, Leisure, News & Politics, Religion & Spirituality, Retail, Science & Technology, Social Networks, Sports, Telecom, Travel, Weather, Others
Year of Collection
2022
Get Started with Off-the-Shelf AI Training Datasets
Appen’s extensive catalog of off-the-shelf (OTS) datasets spans multiple data types and industries, providing comprehensive coverage for various AI applications. These datasets are crafted to the highest standards of quality and accuracy, ensuring reliable training data for AI models.
Talk to an expert