Transcription

The 28 presentations were carefully transcribed in a two-step process. First the entire presentation was orthographically transcribed, including filled pauses. Speech recognition was a helpful tool in the English transcriptions. The speaker-dependent dictation software Dragon NatSpeak 9 was trained to the researcher’s voice, who then repeated exactly what the speakers said into a microphone—a task that required concentration, but made the transcription process very efficient. A complete, though imperfect, transcription could be produced in real time – 10 minutes for a 10-minute presentation. Listening to the presentation two or three more times allowed for correction of the inaccuracies and addition of the filled pauses that the speech recognition is trained to ignore. The vocabulary of the dictation software was impressive, including Swedish place names and rare words such as types of pharmaceuticals and phenomena (e.g. quantum teleportation).

The second phase of transcription, which allowed further correction to any eventual inaccuracies, was to break the transcriptions into ‘runs’, using pauses as boundaries. The speech waveform was used to locate all silent or filled pauses longer than 250 milliseconds. A filled pause is not readily visible in the waveform, and so it was necessary to listen carefully and make run breaks for most of the filled pauses as well, unless they were extremely short ones. The run breaks appear as line breaks in the transcription, but the length of the pauses themselves was not collected as data.