| If I had a hundred pounds for every time | | | | programme and you'll see how difficult a skill that |
| someone told me that transcription services | | | | is to master. Add on the time taken at the end |
| would soon be obsolete and that everyone would | | | | to proofread and tidy up the Word document and |
| be using voice recognition software, I wouldn't | | | | it becomes clear that it would take far longer |
| need to run a business! However, I'm firmly | | | | than the original interview length, and no different |
| convinced that voice recognition software in its | | | | from the time taken by a professional transcriber |
| current form is not a substitute for transcription | | | | to complete the transcription. Is that a valuable |
| services. But then I would say that, wouldn't I? | | | | and productive use of your time? Magnify that |
| Below are some of the reasons why I remain | | | | time and effort many times if you then tried to |
| unconvinced that it will replace me quite yet. | | | | tackle 'speaking' a focus group, especially if the |
| Voice recognition software currently comes in | | | | recording is less than clear. |
| two forms - Dragon Naturally Speaking and IBM's | | | | Another major factor when considering speech |
| ViaVoice. As well as being used to give a | | | | recognition software is its inability to judge which |
| computer commands (e.g. to save a file), it can | | | | homonyms should be used. The latest versions |
| also be used to convert speech straight into a | | | | are apparently now capable of recognising the |
| Word document. However, the main disadvantage | | | | more common ones in simple sentences, such as |
| is that voice recognition software is a 'dog for | | | | deciding whether it's there or their. However, long, |
| one master' only. It's possible to use the software | | | | complicated sentences can defeat it. All this adds |
| successfully for dictation but it can't cope with | | | | to your proofreading time at the end. You also |
| even one-to-one interviews and would probably | | | | need to consider how the software will tackle |
| go into meltdown if you tried it with group | | | | commands for punctuation or formatting certain |
| meetings or focus groups. The software needs to | | | | words in bold, for example. Most programmes |
| be trained to become used to one voice. Asking it | | | | require you to leave a pause between the |
| to recognise and accurately transcribe the | | | | command and the next chunk of 'text'. If the |
| different voice of your interviewee as well as | | | | pause isn't long enough, you'll find those |
| your own questions, or the multiple voices of a | | | | commands entered as part of the text itself, |
| group, is simply not possible at the moment. Even | | | | which will need to be edited out later. For an |
| with one voice dictation, the software still needs | | | | interview format, you will need to indicate a |
| to be corrected and trained to recognise new and | | | | change of speaker with a new line and initials, and |
| unfamiliar words or technical terms or names. | | | | a tab command. Even if you just decide to enter |
| The only way to use voice recognition software | | | | a change of speaker on a new line and tidy it up |
| effectively for an interview situation would be to | | | | afterwards, this all adds to the time taken. |
| listen to the recording and re-speak everything | | | | Allegedly, one of the latest versions (Naturally |
| you hear - both questions and responses. | | | | Speaking) will also punctuate for you - deciding |
| Obviously, this would be very time consuming, | | | | where all the commas and full stops go. Be |
| having to stop and start the recording so that | | | | prepared to correct this later - its idea of |
| you can 'speak' what you're hearing. Essentially, | | | | punctuation is not mine! |
| you're attempting to perform simultaneous | | | | 'Naturally Speaking', as it were, I'm biased! But I'm |
| translation - concentrating on listening to someone | | | | convinced that voice recognition software is a |
| else's speech, while saying those words a few | | | | useful tool fit for a specific purpose, but that |
| seconds later but without losing what's being said | | | | purpose is not yet transcription of all recordings in |
| next. Try it with a recorded TV or radio | | | | all circumstances. |