AssemblyAI Expands PII Redaction and Entity Detection to 47 New Languages
Rebeca Moen Jul 18, 2024 17:35
AssemblyAI enhances PII Redaction and Entity Detection, adding support for 47 new languages and 16 new entity types, improving global data privacy and insights extraction.
AssemblyAI has announced significant updates to its PII Redaction and Entity Detection features, enhancing its Audio Intelligence capabilities. According to AssemblyAI, the update includes support for 47 additional languages and 16 new entity types, making the platform more powerful and globally accessible.
Expanded Language Support for PII Redaction
The latest update to AssemblyAI's PII Redaction feature now supports 47 more languages. This enhancement ensures that Personally Identifiable Information (PII) is safeguarded across various languages and regions, providing robust privacy measures. The feature allows users to securely handle customer service calls, safely share user-generated content, and protect participant privacy in market research.
PII Redaction can identify and remove sensitive data such as addresses, phone numbers, and credit card details from transcripts. It supports both text and audio redaction, ensuring high precision and accuracy. The models achieve over 99% precision, accuracy, and recall in major languages, including English, French, German, Italian, and Spanish.
Enhancements in Entity Detection
AssemblyAI has also enhanced its Entity Detection feature by adding 16 new entity types, bringing the total to 44. This update allows users to extract more value from their audio data by automatically identifying and categorizing key information in transcripts. Entity Detection supports the identification of names, organizations, addresses, and more, providing detailed entity lists and timestamps.
The feature is designed to streamline the process of extracting meaningful insights from large volumes of audio data, making it more efficient and less resource-intensive. It supports various use cases, including analyzing call center interactions, categorizing media content, and extracting trends from market research data.
Entity Detection delivers reliable results with 99% accuracy in major languages and supports EU data residency for 13 languages, helping users maintain regional compliance requirements.
Frequently Asked Questions
Will the expanded PII Redaction and Entity Detection languages be supported by EU Data Residency?
Yes, 13 languages in AssemblyAI's "Best ASR" offering will be supported by EU Data Residency, including English, French, German, Italian, and Spanish.
What is the quality of PII Redaction and Entity Detection across languages?
The highest quality PII Redaction and Entity Detection is found in languages such as English, French, German, Italian, and Spanish, with verified 99%+ precision, accuracy, and recall results.
How secure is my data when using AssemblyAI's PII Redaction and Entity Detection?
AssemblyAI prioritizes data security with enterprise-grade encryption both in transit and at rest. Users can request the deletion of their data at any time, and these requests are handled promptly.
Image source: Shutterstock