To support accurate speech recognition, the following dictation techniques can help improve recognition quality.
Create a dictation-friendly environment
To optimize recognition accuracy, dictate in a quiet, controlled space whenever possible.
- Use a closed office or quiet reading room to minimize background noise.
- Avoid noise sources such as fans, radios, machines, or busy areas.
- Position your microphone away from noise and keep environmental distractions to a minimum.
- Close doors or lower any in-room audio to prevent interference.
- Exit or mute other applications that may generate audio.
Speak clearly and consistently
- Speak clearly and naturally. You can maintain a quick pace as long as your words are well enunciated.
- Use a consistent, normal speaking volume throughout the dictation.
- Keep the microphone at a steady distance from your mouth.
- Avoid actions that may introduce noise into the audio stream, such as throat-clearing, yawning, chewing, or side conversations.
Use dictated punctuation for accurate formatting
- Dictate sentences as you want them to appear in the final report.
- Add punctuation by speaking the appropriate punctuation commands.
- Use colon after section or report headings.
- For punctuation marks that appear in pairs, specify whether it is the opening or closing mark (for example, “begin parenthesis” or “right parenthesis”).
Additional topics to boost and maintain speech recognition accuracy
- Manage vocabulary: Learn to add words, specify pronunciations, and train vocabulary in PowerScribe One for accurate speech recognition and error-free reporting.
- Use AutoCorrect: Explore how to use autocorrect to fix common errors instantly and maintain accuracy in your reports.
- Spellcheck: Learn how to run spellcheck to catch errors quickly and keep your reports clear.