STENO TO VOICE – RESOURCE SHEET (NCRA Annual Convention 2011)
This is not a comprehensive list by any means, but it’s a start for you should you desire to research making a transition – whether due to injury or flexibility. This handout is meant to assist you in starting your research and for you to take notes on during the presentation.
The national organization for voice writing court reporters is the NVRA, National Verbatim Reporters Association. Their website is www.nvra.org.
NVRA promulgates educational materials and seminars, training and certification, as well as advocacy for voice writers.
One of the CORE tenets of the NVRA is to be “the silent voice.” IN OTHER WORDS, you should NOT be heard as you dictate. This is not to say NO sound will come out – some sibilance and perhaps some breathy sounds may be heard, but reporters should NOT be heard as dictating actual words.
Learning to dictate quietly is KEY!
SRE (speech recognition engine)
Dragon or Via Voice (though Via Voice is no longer supported, it is a stable platform. If you can find version 10, it’s a viable choice – if your CAT will support it).
Most people are using the Premium (mid-level) program and getting good results. If you want to go for Professional, it is a much higher price but has some additional functionality. Check with your CAT vendor.
An SRE takes our ‘analog’ voice and digitizes it, and then analyzes the phonemes (sound parts). You must TRAIN the SRE, it is NOT a magic bullet.
Learning how to work with Dragon (or any engine) takes practice – and there are some trainers out there who would LOVE to assist you in making it work. Dessa Van Schuyver out of Atlanta has a training book and holds workshops (at conventions) and user groups locally.
CRITICAL PIECE OF EQUIPMENT: A sound adapter (or ‘pod’ or ‘translator’), the little piece of hardware that takes the analog and turns it into digital sound. When you buy an SRE, you get ONE with your package. Andrea Electronics is the main supplier of this product.
You can start fairly quickly by using the open mic headset (which also comes with your software package) and simply dictate a few stories. Then it’s up to you to upload documents into the software to make it ‘understand’ the way the words in your text files generally appear.
Once you change your input, though, you’ll need to start over and train the system to understand the NEW input device (the closed-mic system of the mask).
MASK (sound-dampening device)
In order to get good recognition, you need to be able to dictate so that the computer can hear you, but the participants cannot. Your main piece of equipment will be your mask. In addition, you should consider other sound-dampening materials such as: foam fillers, mask “wraps” and – believe it or not – nursing pads.
Something to remember about dictating in a closed environment is that you generate a lot of condensation – and the mask is ELECTRONIC. It needs to be aired/dried out several times during the day (I do it at breaks) and cleaned.
There are several different choices out there.
You can get a mask with one or two “tails” – which are your output feeds. If you have a dual feed, you can run one line to your computer and a second to your digital backup recorder. (see below)
You can go high-tech or low-tech. And if you’re using a CAT system already, you may not need to go the high-tech route.
Sony makes one that I use, that has the ability to offload via USB onto your hard drive.
(Sony ICD PX820)
Olympus also makes a similar product. (Olympus WS 700M) Either way, make sure you have a way to offload your files – whether it is a USB cable or an SD/memory card.
Marantz makes many digital recorders that use multiple inputs. These are more expensive and can be used as room backup and, I believe second-output recording.
Of course, anyone who is using a CAT system will get not only their voice track (what could be called “voice steno” feeding in realtime) but will also get their “audiosync” track as well. So any backup system is actually a DUAL backup system. (But I’m a Nervous Nelly and don’t like to have all my eggs in one basket – because if the computer crashes, I’ve got NOTHING. So it’s a two-tailed mask for me with an external backup plugged in)
Learning to Dictate
Some masks cover the mouth and nose; some only the mouth. Figure out what is best for you. Anecdotal data suggests that the over-the-nose mask may lend itself to better translation because of the greater flexibility one has for moving one’s mouth to make the sounds.
Getting a good seal is important, so no sounds come out; however, that means you must learn to dictate without holding your breath for long periods of time. Breaking the seal to breathe, re-establishing your seal, and modulating your breathing is very important.
You can’t just speak regularly into a mask, nor should you whisper. Anyone who has ever had nodes on the vocal cords will tell you that whispering is a no-no. It’s a modulated sound that is your speaking voice, lower and quieter. Hence the need for sound dampening materials.
That said, many people DO whisper. And they get good results. .
Crisp, hard sounds help with word boundaries, so good “D” and “T” sounds will help you if you can slightly overemphasize them in your dictation. Minimal pauses in your formatting macros (say, your Q and A markers, for example) will get a clean recognition.
Be careful with the breathier sounds, like “H” and put emphasis on them. “Him” and “them” are two words that are difficult to differentiate if you’re not careful.
Voice Theory and Briefs
As in steno, the best way to get a good outcome is to use short forms. Each voice writer has their own theories on what works best for them.
However, Bettye Keyes, who has been a voice writer for many years and is one of the few Registered Verbatim Reporters out there (the RVR is the functional equivalent of the CRR) has a text and has many folks who have espoused her method.
What are your high-frequency phrases?
I don’t know? (IDK)
Is that correct? (thak-mak)
(just like in steno, you can tailor just about any phrase into a brief)
Briefing on the Fly
Develop special dictionaries for the unusual words, names
KEYS TO MAKING IT WORK:
- Training your SRE (“feeding your Dragon” as Dessa Van says)
- Editing from your voice file (training your SRE)
- Uploading text from other jobs, pleadings, etc.
- Backing up your user data
- Constantly tweaking your dictation and your voice briefs
- Understanding the difference between what the SRE does and what your CAT does
- Having a top-of-the-line computer (not a netbook, but a real workhorse notebook) and preferably with an SSD (again, check with your CAT vendor for specs)
- Maintaining your mask and pod
- Associating with fellow users to get their input on their voice briefs and other tips and tricks