Systems and methods for automating sensitive data masking using generative AI
Systems and methods for providing interaction recordings with removed personally identifiable information (PII) are disclosed, the systems and methods involving: extracting, from an interaction recording, using an automatic speech recognition engine, a timestamped recording transcript; identifying, using an artificial intelligence (AI) engine, time periods of the timestamped recording transcript which disclose PII; and removing, from the interaction recording, data present during the time periods which disclose PII, to produce a secure interaction recording.
1 . A method of providing interaction recordings with removed personally identifiable information (PII), the method comprising:
identifying, using an artificial intelligence (AI) engine, time periods of a timestamped recording transcript which disclose PII;
comparing text within the timestamped recording transcript within the time periods indicated as disclosing PII to the timestamped recording transcript to identify incorrect PII;
producing PII from the time periods indicated as disclosing PII, the PII not including the incorrect PII; and
removing, from the interaction recording, data present during the time periods which disclose PII, to produce a secure interaction recording.
2 . The method of claim 1 , further comprising:
extracting, from an interaction recording, using an automatic speech recognition engine, the timestamped recording transcript.
3 . The method of claim 2 , further comprising:
dividing the interaction recording into segments before extraction of a timestamped recording transcript.
4 . The method of claim 3 , wherein the method takes place in real time, such that PII is identified in a previously recorded segment, while the interaction is ongoing.
5 . The method of claim 2 , wherein the interaction is complete, and the interaction recording comprises a recording of the entire interaction.
6 . The method of claim 1 , wherein the artificial intelligence engine is a generative artificial intelligence (Generative AI) engine.
7 . The method of claim 1 , wherein the artificial intelligence engine is a large language model (LLM).
8 . The method of claim 1 , wherein the interaction recordings are audio recordings comprising audio data.
9 . The method of claim 8 , further comprising:
replacing audio data present during the time periods which disclose PII with an audible identifier indicative of removed PII.
10 . The method of claim 1 , wherein the interaction recordings are video recordings comprising audio and image data.
11 . A system for providing interaction recordings with removed personally identifiable information (PII), the system comprising:
a memory; and
a processor configured to:
identify, using an artificial intelligence (AI) engine, time periods of a timestamped recording transcript which disclose PII;
compare text within the timestamped recording transcript within the time periods indicated as disclosing PII to the timestamped recording transcript to identify incorrect PII;
produce PII from the time periods indicated as disclosing PII, the PII not including the incorrect PII;
and
remove, from the interaction recording, data present during the time periods which disclose PII, to produce a secure interaction recording.
12 . The system of claim 11 , wherein the processor is configured to:
extract, from an interaction recording, using an automatic speech recognition engine, the timestamped recording transcript.
13 . The system of claim 12 , wherein processor is configured to:
divide the interaction recording into segments before extraction of a timestamped recording transcript.
14 . The system of claim 13 , wherein the system is real-time, in that the processor is configured to identify PII in a previously recorded segment, while the interaction is ongoing.
15 . The system of claim 12 , wherein the interaction is complete, and the interaction recording comprises a recording of the entire interaction.
16 . The system of claim 11 , wherein the artificial intelligence engine is a generative artificial intelligence (Generative AI) engine.
17 . The system of claim 11 , wherein the artificial intelligence engine is a large language model (LLM).
18 . The system of claim 11 , wherein the interaction recordings are audio recordings comprising audio data.
19 . The system of claim 18 , further comprising:
replacing audio data present during the time periods which disclose PII with an audible identifier indicative of removed PII.
20 . A method of removing personally identifiable information (PII) from files, the method comprising:
identifying, using an artificial intelligence (AI) engine, time periods of a timestamped file which include PII;
comparing text within the timestamped file within the time periods indicated as including PII to the timestamped file to identify incorrect PII;
producing PII from the time periods indicated as including PII, the PII not including the incorrect PII;
and
deleting, from the interaction recording, data present during the time periods which disclose PII.