Using the Philips VoiceTracer Software with Dragon NaturallySpeaking

Dragon NaturallySpeaking 9 Preferred editions and higher support handheld digital recorders and other PDAs that record in ".wma" and ".mp3" format for deferred transcription. After a single user dictates into the recorder, the sound file can be transcribed into text when connected to the computer. The Recorder software must be installed first, and then Dragon must be configured to use a Digital Recorder source.

In order to gain the highest rate of accuracy possible, the sound file should be transcribed with a user profile that uses the "Digital recorder using sound files (.wav, .mp3, .wma) on disk" as the dictation source using the steps below.

To create a new user profile using the Digital Recorder source:

1. Launch Dragon NaturallySpeaking 9.
2. Click "NaturallySpeaking > Manage Users".
3. Click the "New" button.
4. Select the "Digital recorder using sound files (.wav, .mp3, .wma) on disk" source from the "Dictation Source" drop down box on the New User Wizard.
5. Enter the name of the new profile and specify the language and vocabulary to use for this new user profile, then click "Next".
6. The New User Wizard dialog box will now state the training steps that need to be completed for the new user profile. Training is not an optional step when using the Digital Recorder source.
7. Click "Next" on the New User Wizard to see the list of possible texts to read. Select the text that will be read, then use the "View.." button to view the text on screen or the "Print" button to print out the text. (Note: When viewing the scripts, the text can be copied and pasted into a word processor.)
8. Once the selected text has been dictated, convert the ".ZVR" file to a ".WAV" file using the conversion steps listed in Part II.
9. Click "Next" on the New User Wizard dialog box, then click the "Browse" button to locate the training file.
10. Once the training file has been selected, click the "Next" button.
11. Now, Dragon NaturallySpeaking will return the results of the Signal to Noise Ratio and the Audio quality checks. Click the "Next" button again to start adapting the user.
12. Click the "Start Adapting" button to begin adapting the user profile to the sound of the recorded speech. (Note: This process may take over an hour to complete).
13. Once the adaptation has completed, the New User Wizard has completed and the new source is ready for dictation.

1. Launch Dragon NaturallySpeaking 9.
2. Click "NaturallySpeaking > Open User".
3. Select the existing user profile to add the new source to.
4. Click "Source > New".
5. Select the "Digital recorder using sound files (.wav, .mp3, .wma) on disk" Dictation Source from the drop down box and then click "OK".
6. Training will have to be completed for the new dictation source. Once the new source is opened, the following message will appear: "Recorder Training has not yet been successfully completed for this user and dictation source. We will give you the opportunity now." To continue with the steps on training the source, please start with step 6 in the directions listed above.

• If a new source is added to an existing profile, vocabularies and commands can be shared between the two sources.
• If a new profile is created that uses the Digital Recorder source, the user profile will use a separate acoustic model than any existing user profiles, and vocabularies and commands won't be shared between multiples user profiles..

1. Click on the ".ZVR" file to be converted in the Download Manager software.
2. Click the "Convert to WAVE" button in the toolbar (see screenshot below).
3. The new ".WAV" file will then appear in the Download Manager software, with the same filename as the original ".ZVR" file.
4. The ".WAV" file can then be opened with Dragon NaturallySpeaking 9 to transcribe a recording. Note: It is not possible to convert to ".WAV" directly from the digital recorder device in the lower pane of the Download Manager. The recorder must be browsed in the upper pane to convert ".ZVR" files to ".WAV" files.

1. Launch Dragon NaturallySpeaking 9.
2. Click "Sound > Transcribe Recording".
3. Select the "Audio File" option and then click the "Browse" button to locate the ".WAV" file.
4. Locate the desired ".WAV" file in the Open dialog box, and then click "Open".
5. Click the "Transcribe" button on the "Transcribe from" dialog box (or click the "Advanced" button first to set advanced settings).
6. DragonPad will launch automatically and the ".WAV" file will be transcribed into the new DragonPad document.

• We recommend using the highest quality audio settings possible for the digital recorder. Please see the Philips device’s user manual or the Dragon NaturallySpeaking 9 Help Files for more information.
• For information on using the Philips VoiceTracer with Windows Vista, please see Technote 5786: "Philips VoiceTracer Software is not able to browse the recorder when using Microsoft Windows Vista".