Representing group interactions

A method of regenerating audio data from a representation of a group interaction, the method comprising:

• extracting, from the representation of the group interaction:
• a conversation path comprising one or more transcript lines, each transcript line being labeled with an identifier identifying a speaker;
• and
• a plurality of voice profiles, each voice profile corresponding to an identified speaker in the conversation path;
• and
• regenerating audio data for the group interaction from the conversation path and the voice profiles.

A system for generating a representation of a group interaction, the system comprising:

• a transcription module adapted to generate a transcript of the group interaction from audio source data representing the group interaction, the transcript comprising a sequence of lines of text, each line corresponding to an audible utterance in the audio source data;
• and
• a labeling module adapted to:
• generate a conversation path from the transcript by labeling each transcript line with an identifier identifying the speaker of the corresponding utterance in the audio source data;
• and
• generate the representation of the group interaction by associating the conversation path with a plurality of voice profiles, each voice profile corresponding to an identified speaker in the conversation path.

A computer program product having a computer readable medium having a computer program recorded thereon for generating a representation of a group interaction, the computer program product comprising:

• computer program code means for generating a transcript of the group interaction from audio source data representing the group interaction, the transcript comprising a sequence of lines of text, each line corresponding to an audible utterance in the audio source data;
• computer program code means for generating a conversation path from the transcript by labeling each transcript line with an identifier identifying the speaker of the corresponding utterance in the audio source data;
• and
• computer program code means for generating the representation of the group interaction by associating the conversation path with a plurality of voice profiles, each voice profile corresponding to an identified speaker in the conversation path.