Scribe4Me: Live transcription for mobile deaf users

Scribe4Me was a mobile sound transcription tool for the deaf and hard-of-hearing. Users could press a button to send the last 30 seconds of audio to a remote worker who would use a desktop interface to write and send back a transcription.

The tool was actually just an extension of Momento, a prototyping platform I built while at Berkeley. Momento included a configurable mobile tool that could capture multimedia and sensor data, and a desktop tool that allowed remote workers to view and respond to incoming data from participants. I built two versions of the Momento mobile tool, one in J2ME and one (pictured here) in C# CF for Windows Mobile. The desktop tool I built in Java.

1

Initially, the transcriber used the Momento desktop interface to configure the mobile interface and add participants to the deployment. Each participant had their own mobile device that was registered in the desktop system. The transcriber monitored incoming requests on a timeline interface.

2

Participants could request a transcript of the last 30 seconds of audio by pressing a button ("what happened?") on the interface. The application saved the audio clip locally and attempted to send it to the server until it was successful.

3

The request would show up on the timeline in the Momento desktop interface. From there the transcriber could listen to the attached audio file and send back a transcription.

4

The Momento platform supported a variety of methods to send data to participants in the field. In this application, the participant received the transcription as a text message.

Tara Matthews ran an extensive study of Scribe4Me in the field.