Project Scope:
We are seeking vendors to manage the Japanese Custom Audio Collection and Transcription project under Project ID SEP-1909(SH433). This project involves collecting spontaneous conversational audio and providing transcription services.
In Scope:
- Access to Japanese-speaking users for audio recording and transcription.
Deliverables & Requirements:
Speech Requirements:
- Speech Type: Spontaneous conversation (no script reading).
- Speakers: 1,000 native Japanese speakers.
- Total Hours: 633 hours of conversational speech.
- Duration: Each conversational speech should average 10 minutes or more (e.g., a 5-minute speech can be included if the total average is 10 minutes or more).
- Recording Environment: Indoor, with minimal noise (clear speech with little background noise, no abrupt interference).
- Recording Device: Smartphone or PC.
- Audio Specifications: 16-bit, 48kHz/44.1kHz/16kHz, uncompressed PCM mono WAV.
- Gender Ratio: 50:50.
Transcription Requirements:
- Format: UTF-8 text format.
- Structure: Tab-delimited.
- Details: Include the start and end time of each utterance.
How to Apply: Please apply via the provided form. Shortlisted candidates will receive further details.