Using Voice Recognition

The voice recognition feature in MMR with M*Modal allows you to dictate reports or report addenda by using a digital voice recorder.

Voice recognition performance improves after the system becomes familiar with your voice and speech patterns. Typically, the system returns more accurate text following 20 minutes of dictation in live mode. This amount of time can be accrued by dictating multiple reports and addenda. In addition, Multi-Method Reporting only learns from signed reports and addenda since the text in a signed report is considered accurate and matching the audio dictation.

For best results when using voice recognition, follow these guidelines:

• Assemble any papers or reports before you start dictating.
• Perform dictations in a quiet room, away from interference and outside noises.
• When dictating, hold the voice recorder approximately 4 to 6 inches from your mouth.
• Wait briefly before starting dictation and wait briefly before stopping dictation. This prevents words from being cutoff when you start and end audio recording.
• Speak clearly and at a regular pace.
• Be consistent in the way you approach similar reports. For example, try to use the same phrases in each of your report types.

The system may misrecognize words due to accent or dictation habits. Reviewing and correcting these mistakes educates the system to perform better voice recognition in future dictations.