Machine Readable Transcripts

The Machine Readable Transcripts dataset aggregates data from earnings calls delivered in a machine-readable format for Natural Language Processing (NLP) applications with metadata tagging.

Leverage Machine Readable Transcripts to keep track of event information for specific companies including dates, times, dial-in and replay numbers and investor relations contact information. Easily combine data from earnings, M&A, guidance, shareholder, company conference presentations and special calls with traditional datasets to develop proprietary analytics.

With this dataset you can rely on:
- Seamless linking between the Speaker ID to the S&P Global Estimates and Professionals database to help identify sell-side analysts' revisions.
- Intraday updates via a feed or streaming XML messages for details on calls scheduled for transcription.
- 11,600+ entities under coverage with history going back up to 2004.

The dataset leverages the power of Kensho NERD to identify company names and aliases mentioned during earnings calls and specifies the mention location in the transcript.

搜集汇总

背景与挑战

背景概述

该数据集提供机器可读的财报电话会议记录，支持NLP应用并包含企业事件元数据，可关联分析师数据库实现修订追踪，覆盖11,600余家企业且数据最早追溯至2004年。其特色是通过Kensho NERD技术识别会议提及的公司名称及位置。