Systems and methods for voice identification and analysis

A method being implemented by a computing system including one or more physical processors and storage media storing machine-readable instructions, the method comprising:

• obtaining configuration audio data including voice information from a plurality of meeting participants in a meeting room, the voice information captured by one or more microphones of a plurality of microphones in the meeting room, each of the plurality of microphones having a respective position relative to each other, the voice information from each participant of the plurality of participants including a respective participant identifier;
• generating localization information based on the configuration audio data and on the respective positions of the plurality of microphones, the localization information indicating a respective location of each participant of the plurality of meeting participants in the meeting room, the respective location of each participant being associated with the respective participant identifier;
• generating, based on the configuration audio data and the localization information, a respective voiceprint for each of the plurality of meeting participants, the respective voiceprint being associated with the respective participant identifier;
• at least during a first time period:
• obtaining first meeting audio data;
• identifying, based on a first weighting of the localization information and on a second weighting of the respective voiceprints, a first segment of the first meeting audio data as associated with a first meeting participant of the plurality of meeting participants and, based on a third weighting of the localization information and on a fourth weighting of the respective voiceprints, a second segment of the first meeting audio data as associated with a second meeting participant of the plurality of meeting participants;
• linking a first meeting participant identifier of the first meeting participant with the first segment of the first meeting audio data;
• and
• linking a second meeting participant identifier of the second meeting participant with the second segment of the first meeting audio data;
• updating the respective voiceprint of the first meeting participant based on the first segment of the first meeting audio data;
• updating the respective voiceprint of the second meeting participant based on the second segment of the first meeting audio data;
• and
• at least during a second time period subsequent to the first time period:
• obtaining second meeting audio data;
• identifying, based on a fifth weighting of the localization information and on a sixth weighting of the respective voiceprints, a first segment of the second meeting audio data as associated with the first meeting participant of the plurality of meeting participants, the fifth weighting being lower than the first weighting, the sixth weighting being greater than the second weighting;
• and
• linking the first meeting participant identifier of the first meeting participant with the first segment of the second meeting audio data;
• and
• generating a first transcription of the first segment of the first meeting audio data and indicating that the first transcription is associated with the first meeting participant, a second transcription of the second segment of the first meeting audio data and indicating that the second transcription is associated with the second meeting participant, and a third transcription of the first segment of the second meeting audio data and indicating that the third transcription is associated with the first meeting participant.