Devices, systems, and methods for audio sanitization
Disclosed herein are computer-implemented devices, systems, and methods for sanitizing an audio recording of a conversation. In an example, such a computer-implemented method can include selecting a transcript of the conversation. The transcript can include a plurality of redaction markers within the audio recording. The plurality of redaction markers can be identifiable via one or more redaction marker start and/or end points. The computer-implemented method can include identifying a first redaction marker and a second redaction marker of the one or more redaction markers. The first redaction marker can be different from the second redaction marker. In this regard, a redaction segment is formable between the first and second redaction markers. The computer-implemented method can include selecting the audio recording. The computer-implemented method can include redacting the audio recording between the first redaction marker and the second redaction marker.
1 . A computer-implemented method for sanitizing an audio recording of a conversation, the computer-implemented method comprising:
selecting the audio recording of the conversation, the audio recording being transcribed to a transcript;
identifying one or more redaction markers at which to mask the audio recording, the one or more redaction markers being identifiable using at least one of the audio recording and the transcript, the one or more redaction markers having a start point and an end point between which a redaction segment is formable; and
sanitizing the audio recording by forming redaction segments at the one or more redaction markers, wherein sanitizing the audio recording comprises:
copying the audio recording from a source file to a destination file up to the point of the start of a redaction segment of the redaction segments;
copying the redaction segment to a memory buffer;
executing an audio masker on the memory buffer; and
writing the memory buffer to the destination file.
2 . The computer-implemented method of claim 1 , wherein identifying the one or more redaction markers includes identifying a plurality of redaction markers to be merged within the redaction segment.
3 . The computer-implemented method of claim 2 , wherein the audio recording is a multichannel audio recording, and wherein when the audio recording is a multichannel audio recording, identifying the plurality of redaction markers includes comparing redaction markers across different channels in the multichannel audio recording.
4 . The computer-implemented method of claim 2 , wherein first and second redaction markers of the plurality of redaction markers are in redtime proximity such that redacting the audio recording between the start point of the second redaction marker and the end point of the first redaction marker minimizes uneven audio.
5 . The computer-implemented method of claim 1 , wherein sanitizing the audio recording includes removing both voice data and background noise that has been recorded in the audio recording.
6 . The computer-implemented method of claim 1 , wherein the one or more redaction markers are identified using the transcript.
7 . The computer-implemented method of claim 1 , wherein the one or more redaction markers optionally correspond to dead spaces within the conversation as indicated by the transcript.
8 . The computer-implemented method of claim 1 , wherein sanitizing the audio recording comprises outputting audio from the audio recording to an audio file such that the audio file includes both masked audio at the redaction segments and non-redacted audio outside of the redaction segments, the audio file thereby becoming a sanitized audio file.
9 . The computer-implemented method of claim 1 , wherein identifying the one or more redaction markers is performed by a machine learning model.
10 . The computer-implemented method of claim 1 , wherein the one or more redaction markers are identified using the transcript.
11 . The computer-implemented method of claim 1 , wherein identifying the one or more redaction markers is performed by a machine learning model.
12 . A non-transitory computer-readable medium that stores instructions that, when executed by one or more processors, cause the one or more processors to:
select an audio recording of a conversation, the audio recording being transcribed to a transcript;
identify one or more redaction markers at which to mask the audio recording, the one or more redaction markers being identifiable using at least one of the audio recording and the transcript, the one or more redaction markers having a start point and an end point between which a redaction segment is formable; and
sanitize the audio recording by forming redaction segments at the one or more redaction markers, wherein sanitizing the audio recording comprises:
copying the audio recording from a source file to a destination file up to the point of the start of a redaction segment of the redaction segments;
copying the redaction segment to a memory buffer;
executing an audio masker on the memory buffer; and
writing the memory buffer to the destination file.
13 . The non-transitory computer-readable medium of claim 12 , wherein the one or more redaction markers is identifiable using the transcript; wherein the one or more redaction markers correspond to dead spaces within the conversation as indicated by the transcript; and wherein sanitizing the audio recording comprises outputting audio from the audio recording to an audio file such that the audio file includes both masked audio at the redaction segments and non-redacted audio outside of the redaction segments, the audio file thereby becoming a sanitized audio file.
14 . The non-transitory computer-readable medium of claim 12 , wherein identifying the one or more redaction markers includes identifying a plurality of redaction markers to be merged within the redaction segment, the plurality of redaction markers including first and second redaction markers; and wherein at least one of:
the audio recording is a mono-channel recording or a multichannel audio recording, and wherein when the audio recording is a multichannel audio recording, identifying the plurality of redaction markers includes comparing redaction markers across different channels in the multichannel audio recording; and
the redaction markers are within redtime proximity to the first redaction marker or the second redaction marker such that redacting the audio recording between the start point and the end point minimizes uneven audio.
15 . The non-transitory computer-readable medium of claim 12 , wherein forming the redaction segment at the one or more redaction markers comprises:
processing the audio recording into redaction time segments and non-redaction time segments; and
outputting sanitized audio to an audio file such that the audio file includes masked audio at the redaction segment.
16 . A system comprising:
at least one processor; and
memory storing computer executable instructions that, when executed by the at least one processor, perform a method comprising:
selecting the audio recording of the conversation, the audio recording being-capable of being transcribed to a transcript;
identifying one or more redaction markers at which to mask the audio recording, the one or more redaction markers being identifiable using at least one of the audio recording and the transcript, the one or more redaction markers having a start point and an end point between which a redaction segment is formable; and
sanitizing the audio recording by forming redaction segments at the one or more redaction markers, wherein sanitizing the audio recording comprises:
copying the audio recording from a source file to a destination file up to the point of the start of a redaction segment of the redaction segments;
copying the redaction segment to a memory buffer;
executing an audio masker on the memory buffer; and
writing the memory buffer to the destination file.
17 . The system of claim 16 , wherein identifying the one or more redaction markers includes identifying a plurality of redaction markers to be merged within the redaction segment.
18 . The system claim 17 , wherein the audio recording a multichannel audio recording, and wherein when the audio recording is a multichannel audio recording, identifying the plurality of redaction markers includes comparing redaction markers across different channels in the multichannel audio recording.
19 . The system of claim 17 , wherein first and second redaction markers of the plurality of redaction markers are in redtime proximity such that redacting the audio recording between the start point of the second redaction marker and the end point of the first redaction marker minimizes uneven audio.
20 . The system of claim 16 , wherein sanitizing the audio recording includes removing both voice data and background noise that has been recorded in the audio recording.