Cobalt's Data Protection Service

How It Works

How Data Protection Works

Cobalt's Data Protection Service is available as a command-line tool that can operates on locally available files. Alternatively, it can run as an integrated part of Cobalt's streaming speech-to-text engine to produce redacted transcripts of your live audio in real-time.

Request Consultation

Redact & Anonymize

Redact and Anonymize

Quickly and accurately redacts 50+ entities of PII in 8 languages with up to 99.5% accuracy.

Redact Audio

Redact Audio

Remove audio from recordings via muting or bleeping to protect your customer data.

Up to 99.5% Accuracy

Up to 99.5% Accuracy

By utilizing the latest advancements in machine learning, we deliver industry-leading performance on semi-structured and unstructured data.

8 Languages

Cobalt's Data Protection Service can accurately detect PII in 8 languages (with more on the way!)

8 Languages
100% Private

100% Private

You retain full ownership of your data, it never leaves you infrastructure.

50+ Entities

Cobalt's Data Protection Service can identify 50+ entities out-of-box, and custom entity types can be added as needed.

50+ Entities

Impressions of working with Cobalt

See how we help hundreds of organizations achieve their goals and increase their effectiveness.

Working with Cobalt Speech and Language technical team was the best choice we could have made.
Megan Davies
Canary Speech

Request a Consultation

Our Latest Posts

Jun 15, 2022
Close up of a woman, face and mouth, with letters floating across the screen. speech synthesis concept
By Rasmus Dall

We’ve previously written about one of our core technologies at Cobalt Speech & Language - automatic speech recognition (ASR). When you speak, the ASR system converts your spoken words into text. Another core technology at Cobalt is text-to-speech (TTS), or speech synthesis, which converts written words into spoken audio.

Jun 3, 2020
3 people talking with each other
By Arif Haque


Many people have used automatic speech recognition systems to transcribe audio to text, but there are a host of other items that it’s useful to identify from a stream of audio. One task in particular is called diarization - who spoke when? Knowing this information can help with a range of downstream applications. For example, in meeting summarization, knowing who said something means you can accurately make notes and allocate action items.