I have spent quite some time looking for a good microphone for Dragon and found there was not much information on the subject.
If you have made the same searches as me, you probably found Nuance’s official recommendations are not useful, and ended up on KnowBrainer’s microphone list ranging from $29 to $475.
While I am not refuting the information from this list, I’d like to show that there are very good, affordable microphones suitable for speech recognition in the translation business.
This precision is important because translating with voice generally requires a quiet place, unlike other professions that may require speaking in noisy places (noise isolation is vital for these).
Here are a few other precisions: I am using a French (France) version of Dragon Professional Individual 14 on a fully updated version of Windows 10 Professional.
So far, I have mainly tried three types of microphones:
Cheap jack 3.5 mm tabletop and headset microphones (~ 5€)
Build quality: bad to OK Accuracy: surprisingly good in quiet environments, at fixed distance Noise cancellation: bad Comfort: usually bad (headsets) Compatibility: no problem with jack connections
Those will work well in a quiet environment, if the microphone remains at the same distance at all time. Dragon is very sensitive to small changes with budget mics. Cheap tabletop microphones are not ideal, but headsets work relatively well in optimal conditions.
It is always useful to have one of these as a spare microphone to carry with your laptop or to have in case of emergency.
Logitech H340 USB Head Set (~ 35€):
Build quality: good Accuracy: very good overall Noise cancellation: good Comfort: OK Compatibility: this headset was not recognized at all by Windows 10 for a while, but it works as good as any other plug-and-play devices now
Works very well. Dragon understands everything I say in a normal quiet environment. Speech recognition works well even when there is small speaker playing radio/music at 30 cm or so, because the microphone is directly before your mouth.
Loud noises (vacuum cleaner nearby or heavy rain on a roof windows directly above your seat) will greatly affect the speech recognition. Then again, using Dragon for translation in such conditions is not the best experience in my opinion.
The main issue with this specific headset is the comfort after wearing it for a while. Upgrading to a higher range headset would easily solve this problem though.
Fifine K669B Condenser Microphone (~ 30€ + ~ 10€ for a boom arm)
Build quality: very good Accuracy: excellent when used properly Noise cancellation: depends on distance from mouth and gain setting Comfort: the best Compatibility: immediately recognized in Windows 10 and DPI 14
This is personally my favorite kind of microphone to work with DPI. When set up on a boom arm, it will not take any space on your desk. The other benefit of this setup is that you can wear headphones while translating with DPI.
Right now, I am using this mic at about 40 cm of my mouth. I just tried using it at 70 cm, without changing any setting, and DPI catches perfectly every single word and command I pronounce.
You might have guessed it already; the main drawback of this microphone is the noise isolation. It is so good at catching sounds that will also pick up other noises when used at such great distances. Therefore, it should be used in a quiet environment, or you need to get the microphone closer to your mouth and to set the gain accordingly.
However, when used in proper conditions, this is from far the best of the three options in terms of performance and comfort.
That’s it for now. This is obviously not an exhaustive list. Besides, I am not recommending these specific microphones, but rather giving you my feedback on three kinds of microphones you can get for less than 40€. I am sure there are dozens of other good options out there for translating with DPI.
To conclude, I would say the operating environment is crucial to determine what kind of microphone you need, and that you do not need to spend hundreds of euros (or dollars) to translate efficiently with your voice.