Real-time contact center speech analytics, with critical call alerts, deployed across multiple security zones

1. A process for operating a telephonic contact center monitoring system, comprising:

(a) at least the following steps, performed within a first security zone:

(i) receiving, in real time, contact center telephony data indicative of multiple agent-caller communications;

(ii) separating, in real time, the received telephony data into tagged utterances, each representing a single utterance spoken by either an agent or a caller;

(iii) using a privacy-filtering automatic speech recognition (ASR) engine to process each utterance, in real time, into a corresponding sanitized ASR transcription;

(b) at least the following steps, performed within a second security zone:

(i) receiving, in real time, the tagged utterances;

(ii) updating, in real time, a database to include each tagged utterance;

(iii) receiving, in real time, a critical call alert, wherein the critical call alert is generated by the system based upon one or more of the sanitized ASR transcription(s);

(c) wherein the first zone has fewer security restrictions than the second zone.

2. The process of claim 1 , wherein the second security zone permits access by fewer users than the first security zone.

3. The process of claim 2 , wherein access to the second security zone is restricted to individuals who have successfully passed a criminal background check, drug test, and credit check.

4. The process of claim 1 , wherein the steps performed within the second security zone further include:

(iv) investigating the critical call alert by retrieving from the database utterance(s) associated with the identified critical call.

5. The process of claim 4 , wherein the steps performed within the second security zone further include:

(v) employing a speech browser to display/play sanitized ASR transcript(s) and corresponding utterance(s) associated with the identified critical call.

6. The process of claim 1 , wherein steps (a)(i)-(iii) are performed without storing any contact center telephony data in non-volatile storage locations.

7. The process of claim 1 , wherein immediately following transcription of an utterance in step (a)(iii), all contact center telephony data that corresponds to the transcribed utterance is removed/whitewashed from any computer readable storage device(s) in the first security zone.

8. The process of claim 1 , wherein the steps performed within the first security zone further include:

(iv) updating, in real time, a database to include the sanitized ASR transcription.

9. The process of claim 8 , wherein the steps performed within the first security zone further include:

(v) providing real time analytics, based on the sanitized ASR transcriptions.

10. The process of claim 1 , wherein step (a)(iii) utilizes an ASR engine to transcribe each utterance and a post-ASR redaction engine redact each transcription in accordance with specified redaction criteria.

11. The process of claim 1 , wherein step (a)(iii) utilizes a privacy-by-design speech-to-text (STT) engine to transcribe only non-sensitive information in accordance with an associated privacy-by-design language model.

12. The process of claim 1 , further comprising an initial step of selecting class(es) of sensitive information to tokenize, including one or more of:

(1) personal names or identifying numbers;

(2) ages;

(3) locations;

(4) organizations or entities; and/or

(5) health conditions, procedures or treatments.

13. The process of claim 12 , further comprising an initial step of selecting one or more of the selected class(es) for stratified tokenization.

14. The process, as defined in claim 1 , further comprising the step of: using a machine learning (ML)/natural language processing (NLP) classifier to identify critical calls, in real time, based on the sanitized ASR transcriptions.

15. A process for operating a telephonic contact center monitoring system, comprising:

(a) at least the following steps, performed within a first security zone:

(i) receiving, in real time, contact center telephony data indicative of multiple agent-caller communications;

(ii) separating, in real time, the received telephony data into tagged utterances, each representing a single utterance spoken by either an agent or a caller;

(iii) updating, in real time, a database to include each tagged utterance;

(iv) using a privacy-filtering ASR engine to process each utterance, in real time, into a corresponding sanitized ASR transcription;

(v) receiving, in real time, a critical call alert, wherein the critical call alert is generated by the system based upon one or more of the sanitized ASR transcription(s);

(b) at least the following steps, performed within a second security zone:

(i) updating, in real time, a database to include each sanitized ASR transcription;

(c) wherein the second zone has fewer security restrictions than the first zone.

16. The process of claim 15 , wherein the steps performed within the second security zone further include:

(ii) providing real time analytics, based on the sanitized ASR transcriptions.

17. The process of claim 15 , wherein step (a)(iv) utilizes an ASR engine to transcribe each utterance and a post-ASR redaction engine redact each transcription in accordance with specified redaction criteria.

18. The process of claim 15 , wherein step (a)(iv) utilizes a privacy-by-design STT engine to transcribe only non-sensitive information in accordance with an associated privacy-by-design language model.

19. The process of claim 18 , further comprising an initial step of selecting class(es) of sensitive information to tokenize, including one or more of:

(1) personal names or identifying numbers;

(2) ages;

(3) locations;

(4) organizations or entities; and/or

(5) health conditions, procedures or treatments.

20. The process of claim 19 , further comprising an initial step of selecting one or more of the selected class(es) for stratified tokenization.