"Because your research questions probably don't include 'what's that weird buzzing sound?'"
Landry Bulls (@LandryBulls)
SCRAP Lab
Created: October 2024
This guide contains information about the audio hardware used in SCRAP Lab for multi-person conversation research. Our setup is designed to record high-quality audio from multiple participants wearing face-mounted microphones during natural conversations. Whether you're studying linguistics, psychology, or just want to capture crystal-clear multi-person audio, this guide has got you covered! 🎯
Our setup is designed to capture isolated, single-speaker audio from each participant in close-quarters conversation, saving everything as perfectly synchronized .WAV files to an SD card. The entire setup can be assembled into a portable rack that can be carried or rolled around (think of it as your mobile conversation-recording command center! 🚀).
While we use the wireless version described below, a more budget-friendly "hard-wired" setup is possible if participants will remain stationary. Here's what you'll need:
- Option A: Countryman E6 (Untested but promising and less expensive)
- Option B: DPA 4088 (What we use but is probably overkill)
You'll need a receiver system with:
- Multi-channel receiver unit
- Bodypack transmitters (one per participant)
- Recommended: Nady 4w-1ku
Perfect for conversation research without the overkill of professional theater equipment!
Choose based on your participant count:
- For larger groups: Zoom LiveTrak L-8 (Supports 8 participants)
- For pairs/mobile: Zoom H6
- TRS Cables: Hosa TRS cables (One per channel)
- Rechargable AA batteries and charger for the transmitters
Pro tip: Get well-shielded, short (1-2 ft) cables to minimize interference!
- Storage: SanDisk 128GB SD card (Have extras!)
- Adapters: Various mic-to-receiver adapters
- Power: SurgeX Power Conditioner
- Optional but recommended: Gator Cases Rack
graph LR
A[Microphones] -->|Audio| B[Transmitters]
B -->|Wireless| C[Receiver]
C -->|TRS| D[Recorder]
D -->|Digital| E[SD Card]
- Each transmitter must be set to match its corresponding channel on the receiver
- Based on our testing, FM channels should be spaced at least 16 channels apart to prevent cross-talk and interference
For example, with 4 participants, you might use channels 1, 16, 32, and 48 to ensure clean separation
Need to clean up your recordings or sync with video? Check out:
- Audio isolation script for isolating individual microphone channels for enhanced separation
- Simple transcription with WhisperX for using WhisperX to transcribe your audio files
- Functions in this script for aligning audio with video files (you will need to adjust the code to fit your specific setup).
- Always test your wireless setup before recording
- Keep spare SD cards handy
- Monitor battery levels in transmitters
- Consider ambient noise in your recording space
- Use good cable management to prevent interference
Happy Recording! 🎉
For questions or contributions, feel free to open an issue or pull request!