## UPDATE ## This review was initially for the LFH0660/10 – this has been replaced in Australia in 2010 by the new model LFH0617/00. The software is the same, the recorder included in the pack is different. The Dragon NaturallySpeaking DVR (Digital Voice Recorder) Edition now contains the LFH0612 Philips digital voice tracer. The below review is still relevant, the software is setup and trained the same way with the new pack.

Released mid 2009 in Australia in a joint venture between Philips Dictation and Nuance was the Philips Dragon NaturallySpeaking Voice Recorder Edition (LFH0660/10) at an incredibly cheap price of RRP AU$259.00 incl. GST. At that price it certainly makes you sit up and pay attention, after all the package contains Dragon NaturallySpeaking v10 and a very competent Philips Digital VoiceTracer 660.

Not much to say about the Philips DVT 660, a great little recorder, perfect for single speaker audio. The 660 records in MP3 format and like all recorders has multiple recording modes (tip: always use the highest quality recording mode, SHQ on the DVT 660). This is a note taker not a dictaphone so you are not able to rewind/review your audio just simply record and stop, although you can pause during recordings. The recorder has 1Gb of in-built flash memory, can be powered by two AAA batteries or by USB and can record on the highest quality audio setting for 17.5 hours.Included in the box is version 10 of Dragon NaturallySpeaking (Windows only) which uses the latest Dragon speech recognition engine, this is a drastically cut down version of the software so does not include features that you would see in the Standard or Preferred edition, hence the cut down price. The recognition engine though is the same so no skimping on the quality of the speech recognition.

The aim of the  Digital Voice Recorder Edition is to allow people to dictate into their digital voice recorder for later transcription by the Dragon NaturallySpeaking software. The user does not need be in the office and in front of their computer while dictating, therefore gives them the freedom to record dictation whenever and wherever they want, which of course is very convenient. Because the software is a cut down version you can not directly talk to the software using a headset and mic as you can with the Standard or Preferred versions, the software will only transcribe audio from a digital recording. Now this does not have to be a recording made on the 660, any recording in the audio format .mp3, .wav or .dss can be used. The software can be configured for multiple user profiles (or speakers) so if you have a number of people in the office all with a digital voice recorder that records in a compatible audio format they could all use one installed copy of the software, once they have trained their profile to their voice.

Installing Philips Dragon NaturallySpeaking DVR Edition

Installation of the software is quick and easy and you are guided through each step of the install with clear and precise instructions, normal with any Dragon NaturallySpeaking install. When installation is complete you are presented with a selection of eight transcripts, any one of which you are required to read for your initial voice training. To do this read your preferred transcript and record your voice using the Philips VoiceTracer 660. This must be recorded into one single digital audio file and not split over multiple files. You have the option to display the transcripts on the screen or print them.

The selection of eight voice training transcripts to read are:

  • pdf-file-logo-icon3001 – The Final Odyssey (Harder Reading – Science Fiction)
  • pdf-file-logo-iconAlice’s Adventures in Wonderland (Reading for Teens – Fantasy)
  • pdf-file-logo-iconCharlie and the Chocolate Factory (Reading for Teens – Adventure)
  • pdf-file-logo-iconCharlie and the Great Glass Elevator (Reading for Teens – Adventure)
  • pdf-file-logo-iconDave Barry in Cyberspace (Medium Reading – Humour)
  • pdf-file-logo-iconDogberts Top Secret Management Handbook (Harder Reading – Humour)
  • pdf-file-logo-iconSuccess is a Journey (Medium Reading – Business Philosophy)
  • pdf-file-logo-iconThe Captain of Battery Park (Reading for Teens – Adventure)

I suggest you print one then take yourself away somewhere comfortable to sit and read the transcript while recording your voice with the digital voice recorder.

Voice Training is done by reading a transcript into your digital voice recorder

Tips when recording your first voice training transcript:

  • Speak in your normal voice but make sure your speak clearly.
  • Speak at your normal speech rate, do not rush or speak too slowly.
  • Make sure the digital voice recorder is set to SHQ recording mode.
  • Take some time to understand how the voice recorder works before you begin. You can pause during the recording of reading the transcript, work out how to do this before your start.
  • Record the audio somewhere quiet, not at a cafe or in a busy office with phones ringing and people talking around you.
  • Have a glass of water handy and take some sips if and when your pause recording, helps keep your audio clear.
  • Do not dictate punctuation.

I was interrupted a couple of times in my office while recording my voice training transcript hence the tip for learning how to pause the recording. Training the software to recognise your voice and more importantly the way you speak is key to the success of this software and ultimatly your user experience with it. Be warned the selection of transcripts can take some time. For my voice training I selected “Charlie and the Chocolate Factory” which took just over twenty minutes to record (with a couple of interruptions).

Once you have completed the recording of your voice training audio connect the digital voice recorder to your PC via USB. Then select the location of the audio file which by default is in Folder A on the voice recorder unless your specifically selected another folder prior to recording. By default the software looks for a file called in.wav, simply browse to the voice recorder and select the recording you just made:

Select your voice training audio on your digital voice recorder

Dragon NaturallySpeaking will transfer a copy of the audio from the recorder to a temporary folder on your PC where it will verify the audio. Verification checks the audio quality and clarity of your recording as well as the signal to noise ratio. If all is good Dragon NaturallySpeaking will identify which one of the eight transcripts you chose to read, as you can see from the image below “Charlie and the Chocolate Factory” was identified, click next and the real work begins. Dragon will process your audio against the transcript and work out how you speak and how you sound.

Dragon NaturallySpeaking verifies your audio for clarity

Verification of the audio can take up to an hour, for me it only took around 20 minutes until the process was complete. Nearly finished, hang in there…. Finally you have the option of having Dragon NaturallySpeaking analyse your writing style, the software can look at your email (Microsoft Outlook, Outlook Express or Lotus Notes) and your documents (Microsoft Word, Text and Rich Text files) and adapt to your writing style. This helps Dragon NaturallySpeaking produce documents in your style of writing, this part can be skipped.

That completes the voice training and initial setup for Dragon NaturallySpeaking Digital Voice Recorder Edition. It is recommended that when you have time in the future you work your way through the other seven voice training transcripts, that helps Dragon fully analyse the way you speak and enunciate your words.

You are now free to record dictation to your Philips DVT 660 and request Dragon to transcribe the audio for you. You will need to include punctuation as you dictate, “new paragraph”, “new line”, “full stop” and so on otherwise your transcript will be one long sentence in one long paragraph. I found this out when I decided to get Dragon to transcribe my voice training transcript, below is a screen shot of Dragon NaturallySpeaking in action transcribing my training dictation:

Dragon NaturallySpeaking transcribing audio recorded on a Philips Digital Voice Tracer 660 - LFH0660

You will still need to proof your transcribed audio however the whole exercise will still be alot quicker than typing the audio yourself.

Remember, this software only works for one person dictation. You can not record an interview or meeting and have Dragon transcribe it for you. You should not even attempt this as other voices will change your voice profile and affect the accuracy of transcription. For multi speaker audio you need your own on-site transcriptionist or outsource your digital audio to Australia’s leading online transcription service – The Transcription People.

So to sum up, I was actually very impressed by the ease of use of the software and the transcription accuracy. I am very used to using Dragon NaturallySpeaking Preferred using a headset and mic so just recording into a voice recorder put me a little out of my comfort zone but for people who are used to talking into a voice recorder they will be very comfortable with the whole process. There is minimal effort to proof your completed transcription and the speed of transcription will easily beat the fastest typist.

Pros:

  • Cost effective, very cheap for excellent software and a quality voice recorder.
  • Easy to use software and digital voice recorder.
  • Fast, accurate transcription for single speaker dictation.
  • Software can have multiple user profiles so anyone with a voice recorder can have their dictation transcribed.

Cons:

  • Document has to be proofed post transcription and not in real time as with Standard or Preferred Editions.
  • Voice training audio can take up to twenty minutes to record.