# RadioTalk: a large-scale corpus of talk radio transcripts

## TL;DR

RadioTalk is a comprehensive large-scale corpus of US talk radio transcripts from 2018-2019, designed to support NLP, social sciences, and conversational research with extensive metadata and analysis.

## Contribution

This paper introduces RadioTalk, a novel, large-scale, publicly available corpus of talk radio transcripts with detailed metadata and initial descriptive analyses.

## Key findings

- Corpus contains 2.8 billion words from 284,000 hours of radio.
- Includes metadata such as location, gender, and program info.
- Provides initial descriptive statistics and high-level analyses.

## Abstract

We introduce RadioTalk, a corpus of speech recognition transcripts sampled from talk radio broadcasts in the United States between October of 2018 and March of 2019. The corpus is intended for use by researchers in the fields of natural language processing, conversational analysis, and the social sciences. The corpus encompasses approximately 2.8 billion words of automatically transcribed speech from 284,000 hours of radio, together with metadata about the speech, such as geographical location, speaker turn boundaries, gender, and radio program information. In this paper we summarize why and how we prepared the corpus, give some descriptive statistics on stations, shows and speakers, and carry out a few high-level analyses.