Having worked in transcription for over a decade, I have heard just about every transcription myth that exists. After hearing them over and over again, I decided that it is time to demystify some of these transcription myths.

1. It takes an hour to transcribe an hour long audio recording

This could be the most popular transcription myth that floats around, but it does not hold any water. It can take longer than an hour, even up to three hours or longer, to transcribe an audio file. However, this depends on some aspects including:

• The technical nature of the recording
• The accents of those captured in the recording
• The client’s formatting needs
• The transcriptionist’s general experience

Another thing to have in mind is that audio recordings with multiple speakers generally take longer to transcribe. For instance, an hour long audio recording with multiple voices could take between 3 to 4 hours to transcribe.

The transcript also goes through a confirmation stage involving listening to the recording once again. This stage takes an equally longer time than the actual length of the audio recording.

2. Transcriptionists need to be fast typists

This is one of the other myths commonly floating around about transcription. It is in actual sense not true; in a s much as fast typing skills are an added advantage when it comes to transcription, what matters most is a keen ear to get down the right information.

The first step is to have a listen at the recording and to accurately type the information.

It seems like an obvious step right?

Surprisingly, this is the most difficult phase of transcription and even the clearest recording still takes time to transcribe.

Dealing with a recording with several voices is often the toughest task, as well as working on a recording with several voices with diverse accents. It can also be a difficult task to carry out transcription for a new client or on an unfamiliar subject.

Take note of the client’s requirements and guidelines to carry out the transcription effectively.

Clients have differing needs and it is important to accurately carry out transcription according to the client’s requests. Formatting requirements, whether verbatim is needed or not and the turnaround time are important aspects to note down when initially taking on the task.

3. Transcription is simply typing, anyone can do it – it’s easy!

The number of times I have heard this could make me a millionaire if I got a dollar each time.

It is true that technology has advanced and most people have computers at home, and pretty much everyone can type. However, not everyone can transcribe; transcription involves a combination of listening skills, hearing and documenting what you have heard. As easy as it sounds, it definitely is not that easy and that is why we are not all transcriptionists.

Transcription is also not everyone’s cup of tea and most people view it as an arduous task, hence the reason why most people outsource this work to others. The work also involves a good level of research to make sure that the names of people, places and terminologies are documented correctly.

One can customize their transcripts and there are options that can be discussed by Transcription wave.

Indiscernible

Transcription wave mark the indiscernible parts of transcription, these are parts that cannot be heard or are not definite as “[indiscernible].” Before declaring it indiscernible, the transcriptionist goes through it three times just to make sure. This is done in order to bring down your costs.  But in instances where the section is very important then more time will be spent on these parts or by listening to the tape again.

In the very hard cases, we can digitize and filter the audio for parts that must be recovered. We can filter out the background noises covering up the conversations. With this the material can be recovered. This brings up the cost because of the recovery though.

Verbatim

One can decide to get the verbatim or edited transcript depending on customer request. With verbatim, the transcripts include false starts, repeated words, stutters and does not correct grammar. It does not include the ‘ums’ and ‘ahs’

Edited transcription

With edited transcriptions, the spoken English is converted to written English, grammar is corrected, and it has no false starts, repeated words or stutters. It does not include ‘ums’ and ‘ahs’.

Guessing

The transcriptionist will not guess the words said, but will just use the conversation to get the context of the sections in order to understand them better.

Grammar

Grammar can be corrected as we transcribe, that is if requested. Spoken English is not the same as written English and may have errors. You can decide whether you want this done to your transcription. You can also choose to receive you transcription in verbatim where mistakes will not be corrected. More on verbatim transcription

Getting to the cost of transcription can be varied. It can be determined by the material, the time taken to transcribe, piece rate or bid by the job as per customer request. Other determining factors include quality of the recording, editing required and technical content .Not every service provider will cost the same way. Transcription wave considers all these and determines what will be the best for you as a customer.  However, depending on the material and the consistency of the audio files and tapes, only a particular method can be used transcription.

The following information gives details that may determine transcription cost.

Length of Audio & Transcription time

Several things determine how long it takes to transcribe. It can take between two to six times the lengths of the audio. The length mainly depends on type of material, pace of talking, quality of audio, how many speakers are there and how clear they are. Transcription wave takes about 2.5 to four times the audio length. Transcriptions that have only one speaker or are interviews that have clear audio take less time.

One can choose to pay per hour of labor. We have professional typist who type fast therefore reducing the hours.

Length of audio and transcription pages

In single speaker audio, the approximate range of transcription is one page per minute. An average page for most transcription services has 22 lines with 65 characters that are single spaced across. Transcription wave has 25 lines per page and this will cut your transcription bill by 12%. With an interview or one speaker transcriptions they may amount to less that one page per minute. As for multi- speaker audio and fast speakers, it may become more than a page per minute.

Technical content

This is another factor that determines cost of transcription. Also the amount of editing and technical knowledge required is a cost determinant. Some cost more like medical transcriptions. This is because they need more skill, tools and references; these are needed to ensure that the transcript is logical and usable.

Cost is also higher in business transcriptions as they require a lot of grammar corrections. The amount of time taken by your staff for special requests, rush requests and inquiries will also determine the cost.

As for legal transcriptions they are quite similar to business transcriptions, this is because court proceedings are varied in content. Costing is done per page in such jobs as it shows the work and transcription involved clearly.

Verbatim & Editing Transcription

This factor can add or reduce your costs. Transcription wave edits many documents as they transcribe them. This means that the end product is not verbatim and it will require little editing once it finished. This will reduce ones cost as charging will not include the lines, words, pages that are edited out. The editing is included in the total cost. As a customer, you may require some documents in verbatim which Transcription wave do.

Not all companies edit and they provide only verbatim transcripts, in the end this will end up costing you more. This is because you are also paying for what will end up being edited out. That will mean that have to edit it yourself or do it yourself. Transcription wave transcriptions will cut you these costs as they have secretarial skills.

Costing per line or word

You have to know what a line is when you are considering the cost of words verses the lines. In transcriptions, a line is not determined by the lines in your original work. It is defined as the number of characters across. Some, like medical transcriptions, are charged per page and interviews are determined by pages. However this is not fixed and can be converted if requested.

Paying by line and word can be a good deal. However some documents may have short lines and will be more expensive if the rate is done per page, line or minute rather that per word. Paying per line requires careful consideration as you will end up paying for empty space. Documents like this are best charged per word. This is good as it can be easily verified and billing is made easier.

Cost Per Minute

Many transcription services offer per minute rate. Most consumers go for this as the cost can be quickly easily determined even before you hand over the raw content. There is a catch though as one may end up paying more.  You may have a slow speaker who will increase your minutes. In this instance you may prefer per page rate. This rate is better for verbatim transcription rather than edited work.

Conversion summary

Here are some of the rate conversions that can help you understand the charging of the transcriptions. These are only approximations and may vary for your documents.

Pages (standard page*) & audio length

• 0.5 to 0.75 pages / audio minute for slow speakers, some single-speaker presentations
• 1 page / audio minute for  single speaker regular dictation
• 2 to 4 pages / audio minute for multiple speakers or fast-paced dialogue

Lines per page (standard page*)

• 22 lines / page  (normal)
• 25 lines / page (Transcription wave – 12 % savings)

Hours labor to transcribe audio

• Usually 2.5 work hours per audio hour to 4 work hours per audio hour
• Up to 6 work hours per audio hour for difficult audio (that requires listening multiple times)

Words per line (standard line of 65 characters including spaces)

• About 8 to 9 average words per line depending on number of blank lines in document

Characters per word

• About 5.5 to 6 characters/word on average, depends on word complexity

Standard Page

• 65 characters (including spaces) per line (usually 12-point mono-spaced font)
• 22 lines per page (Transcription wave uses 25 lines per page – 12 % savings)

Rush transcription

Cost can also be determined by the turn around on a rush basis. Transcription wave is available for rush services that can go to up to 24 to 48 hour. For regular and high volume clients standard rates will be kept standard for rush work.  However, for unscheduled rush work cost will be higher than standard rate.

Transcription wave are at hand to give rushed services depending on customer needs and the company’s ability at that time.

Be cautious of other companies that have unlimited rush offers. This is because the work may end up being done outside the country and therefore reducing the standards of your work.  To avoid this situation, test them first by sending work that you are really not in a rush with.

Best cost to transcribe

To find the best cost you can contact Transcription wave and you will get to figure out the rate you will be charged for your work. It is also good to sample other companies; you can get their rates by sending a sample of the work you want to be done. This will help you decide the best way to go. More on transcription services cost

Other than human labor, it is also possible to transcribe speech into text using Adobe premiere CS4 & sound booth CS4. In addition to this, motion graphics fire power can also be used with the after effects CS4. Speech can be transcribed from a Quicktime video to come up with the subtitles needed. To get this done, you will need:

• Adobe Premiere Pro CS4
• Barackin.mov

Creating transcription in Adobe Premiere CS4

Start a new project:

1. Begin the Adobe Premiere Pro
2. Select New Project from the Welcome Screen
3. Specify the name and location of the project in the New Project Panel
4. Click OK
5. Select DV- NTSC Standard 48 kHz in a New Sequence dialogue box
6. Click OK

Importing the video clip

1. Select file > import then move to the BarackIn.mov clip
2. Import the video file to the Premiere by double clicking on it
3. Double click on it once it is on location to have it open at the Source Monitor
4. Use transport controls in the Source Monitor to view the video for a few seconds. Pay close attention to the audio when doing this to judge whether a good job will be done at the end.

Open MetaData View

1. Select the Metadata window to see whether it is enabled
2. Check to see whether it has 3 headings i.e. clip, file and speech.

Creating the Transcription

1. Ensure the clip has been selected and highlighted on the project panel and proceed to click on the transcribe command.
2. You will get a number of options where you are supposed to set the appropriate setting according to the content of the video
3. Select between high quality processing that is usually slower and medium quality processing that is normally faster.
4. You can select identify speakers if you are dealing with more than one speaker.
5. Click OK
6. Adobe Media Encoder handles transcription in Adobe Soundbooth and Adobe Premiere and it launches automatically when you have completed the transcription dialog.
7. The clip should be set in the correct format when it is loaded into Adobe Media Encoder
8. Click on Start Queue command
9. The Encoder encodes the clip while adding the speech transcription metadata on to the video file.

You can work with the new transcription Metadata to correct some of the mistakes made in the transcript.

Transcription services have now been around for long periods of time in one form of another just like written language to capture the essence of momentary speech. This is because transcriptions deal with the writing down of speech as it is happening. This however does not mean that the services have been as accurate and thorough as they have been the past couple of years.

Transcription was mostly done by hand before the introduction of reliable means of mechanical and note taking techniques. This means that people used pencils, pens and other manual writing tools. This should tell you something about the quality that was produced. It is quite obvious that writing is much slower than talking save for a few special cases. This means that when using long hand, it is quite impossible to write what someone is saying accurately.

At one time, this was solved by the creation of shorthand which is still used to date mainly when note taking and explicitly manual transcription is the only course of action that can be taken. People who practice short hand also suffer from the same problems as long hand in terms of speed and mechanical inefficiency.

Transcription however did not reach its full potential until the type writer was introduced in the market.  This allowed a person to write much faster than short or long hand making it one of the most superior methods to capture speech back in the day. It however posed some problems when it jammed and needed to be refreshed so that it could work but all in all it was a really great invention. From here computers were discovered and they replaced the type writers as they were more efficient and flexible.

The ability to write mechanically however was not the greatest invention when it came to transcription. This is because recording of voices was. This however did not make an appearance until the seventies when the tape recorder could be used. This allowed individuals to record what someone had to say and play it back over and over again which was very useful for transcription services.

Despite the fact that there have been numerous technological advances that have helped with transcription services, it is important to make sure that you look for reliable and affordable companies that offer the services so that you can get quality work done on time. This is because it does not matter how great the technology is until the experts come up with software that can be used to automatically and accurately transcribe speech.