Take Note Team
Have you ever tried to convert a video or audio passage into text? Doing it on your own can be a laborious and painstaking process. There are many professional or academic contexts in which you may require regular transcription, and keeping up with it on your own can be a colossal drain on your time and energy.
Fortunately, the digital era affords you access to a wide range of transcription services to help ensure that you’ll never have to attempt this arduous chore on your own again.
Choosing a transcription service, however, is more complicated than you may think. You have a wide range of human and automated transcription services to choose from and it’s important to know your needs as well as the capabilities and limitations inherent in each type of transcription service.
Transcription on the go
Record & upload files instantly on your mobile and get them transcribed using our app!
The Goldilocks conundrum — Finding the transcription service that’s just right
There are many factors which need to be considered when choosing a transcription service. Perhaps the most fundamental is whether you opt for a human-based transcription service or an automated application that uses ASR (automated speech recognition). Both have their benefits and caveats as we will shortly discuss.
Cost is another important consideration, but keep in mind that the cheapest service may not be the best option for you. Security is also an important consideration, especially when transcribing potentially sensitive information.
There’s also the matter of flexibility and detail. Does your transcription service offer access to different kinds of transcripts? Because not all transcriptions are created equal.
Here we’ll look at some of the most important things to keep in mind when choosing the best transcription service to suit your needs...
Know your accuracy standards and accuracy options
On the face of it, ASR may seem like the perfect option to service your transcription needs. It’s fast and affordable and easy to access with prices averaging around 7-10 pence per minute. Nonetheless, the savings you make in time, effort and money may be offset by a lack of accuracy.
Currently, although ASR solutions and the algorithms that power them have become markedly more sophisticated in recent years, they still max out at 80% accuracy. While this sounds impressive it can lead to a challenging false economy. While ASR software might be able to transcribe an MP3 file in seconds, you may nonetheless have to spend valuable time proofreading your transcript to check for homophones or words which are just plain wrong.
Unlike a human transcriber, an algorithm can’t determine the right word or spelling dependent on syntax and context. Nor can they account for different accents (hands up Scottish people who are frustrated by Siri?) or differentiate vocal tics like laughter, stammers or false starts from speech.
It’s also worth remembering that 80% accuracy is a “best case” figure. Poor quality audio, a lot of background noise, multiple speakers and different accents may significantly reduce this figure.
While human transcribers may be a slightly more costly option with prices varying from 50p to £2 per audio minute, they offer a far better standard of accuracy and versatility. They are especially appealing to those with loose timeframes. You can expect to pay a premium if you need your transcription done quickly. However, if you have a few days to spare, you can get the best rates, ensuring that you get 99-100% accuracy at an eminently reasonable price.
Know your transcript options and choose the level of detail you need
Different professions and academic disciplines have different transcription needs and there may be some circumstances under which they can only be met by a human transcriber. Sometimes you may want the audio recording in its rawest form, and need as much detail as possible including pauses, false starts, stammers, “ums”, “ahs” and other vocal tics and mannerisms. Other times you may not need a full transcript of the audio at all, as it is too lengthy and difficult to parse in which case a set of notes would actually be more useful to you than a full transcript.
The great thing about human transcribers is that they can offer you flexible solutions including:
- Full Verbatim — Full verbatim is the raw transcription. It includes a full and accurate transcription of the audio content as well as every single pause, vocal tic and filler sound. At a small additional cost, you even request detailed notes on these.
- Verbatim — Verbatim or “intelligent verbatim” is a more polished version of the speech. While it retains the same level of accuracy and doesn’t paraphrase, it edits out pauses, tics and filler words.
- Detailed Notes — If the essence of the speech is more important than the speech itself, you may find that detailed notes are the most appropriate option to suit your needs. Detailed notes cover all the main talking points with off-topic fluff removed. This is perfect for those who want a quick reference as opposed to a daunting chunk of text they have to review themselves for the relevant information.
- Extras — If you want extra clarity and detail, you can add extras for a small additional cost, such as speaker identification, accent identification and time stamps for quick reference.
Know what makes a secure transcription service
Unless you’re transcribing data that’s already in the public domain, security is an extremely important consideration when choosing a transcription service. If you are a business looking for a transcription service, you should consider your relationship with your service provider as important as your relationship with any other vendor. The integrity of your business depends on your ability to trust them.
Here are some signs that your chosen transcription service can be trusted to handle your data in a secure and ethical way:
- They’re local and therefore accountable to local data protection laws.
- They have a secure https web address with encrypted portals and storage.
- They’re ISO accredited. Look for ISO 27001 and 9001. This means that they have undergone a rigorous formal assessment by a third-party when it comes to their information security management.
- They’re willing to sign a Non-Disclosure Agreement (NDA) to make themselves legally accountable for the handling of your data.
One might assume that, because there is no human involvement, ASR solutions are inherently more secure. However, that isn’t necessarily true. Developers may have access to your recordings and use them for quality control purposes. What’s more, they may outsource their testing to third-party companies. In this case, you will have no way of knowing who has your data, where it’s being stored or how many copies of it have been made.
Whether your chosen solution is human-based or automated, you deserve to know exactly who has access to your data.
Know which one is right for you!
Ultimately, there is no right or wrong answer. It’s simply a matter of determining which transcription services best suit your needs. Some may consider staying under budget a priority and these people may be better suited opting for an ASR solution and putting in the time to correct inaccuracies themselves. On the other hand, for others, accuracy may be the key, and any inaccuracies can either be damaging to their thesis or costly to their enterprise.
With the market for transcription services becoming increasingly widespread and diverse, this has led to an ever-expanding matrix of complex needs. As such, the wealth of different automated and human-based transcription services may seem daunting at first but can actually help to put users in the driver’s seat.
With just a little research, you can choose the service that’s best aligned to your specific needs, budget and priorities!
You’ve been reading about how to select the best online transcription service — we hope this article has helped you. If you want more information about choosing a transcription service that’s right for you, our Ultimate Guide to Transcription Services is perfect for you.