Lyra Diarization Project

Description :

This is a remote transcription project focused on diarization (the identification and separation of individual speakers in audio clips). Participants will label speech segments, identify human speech, assess conversational context, count speakers on an audio segment, and mark speech boundaries.

Purpose :

To ensure each voice in audio data is clearly identified and faithfully represented, contributing to a dataset that will support the development of next-generation speech technology.

Main Requirements :

• Native or fluent speaker of one of the following locales : French (France), German (Germany), Italian (Italy), Spanish (US), Spanish (Spain).
• Ability to distinguish and label human speech in audio
• Capable of evaluating conversational context and marking speaker boundaries
• No app download required