US English Speakers - Smartphone Voice Study

Shaip is a global leader in AI data solutions, specializing in sourcing, annotating, and managing high-quality audio, speech, video, and text data for machine learning models. We help technology organizations scale their AI initiatives by partnering with structured crowds and specialized human-in-the-loop work forces worldwide.

Please see the full project specifications below:

PROJECT OVERVIEW
We are looking for native American-English speakers across the United States for a short, remote voice recording study. The recordings will be used to train and test voice-recognition software for wake phrases (e.g., "Hey Siri" or "Hey Google"). This is a one-time task completed at home on your smartphone, taking about 20–30 minutes.

WHAT YOU WILL DO

• Record 48 short phrases: 30 versions of a provided phrase and 18 similar-sounding phrases.

• Say each phrase at three speeds: slow, medium, and fast.

• Record in a quiet room at home using your phone.

WHO CAN PARTICIPATE

• Adults 18 and older living in the United States.

• Fluent American-English speakers whose everyday speech reflects their US region (Northeast, South, Midwest, or West).

• All genders and age groups are welcome. Participants complete demographic questions at sign-up to ensure a balanced mix.

EQUIPMENT AND SETUP

• Use your phone's built-in microphone only (no external or studio microphones).

• Find a quiet room without background noise or strong echo.

• Place your phone on a stable surface or stand (not handheld) at your assigned distance:
  Near (1–3 ft), Mid (6–10 ft), or Far (13–16 ft).

• Speak naturally and clearly, pausing about half a second before and after each phrase.

PAY

• $12 USD flat fee paid within 7–10 business days after recording approval.

• Recordings are verified for clear audio, accurate reading of phrases, correct distance setup, and regional match.

• There is no cost to participate. We will never ask for payment or your Social Security number.