When you think of a stenographer, you probably picture someone sitting in a courtroom, frantically typing out everything that’s being said, but this is just a simplified version we see on TV.
In reality, it’s more complicated than that, and once people start using words like typist, palantypist, and transcriptionist, then things really start to get confusing. Read more →
You might think we only provide remote captioning services, but we actually also provide captioning services on-site in the UK and Europe. Remote and on-site services have their pros and cons; which service will work best for you very much depends on your individual situation and preferences.
Speech-to-text reporters are stenographers or palantypists who use a shorthand machine to produce a verbatim real-time translation of what is being said, with a minimum accuracy of 98%, unlike notetaking and respeaking.
Having a captioner on-site might work better for you in some situations than having them work remotely. Here are some of the benefits of on-site captioning services provision.
An on-site speech-to-text reporter needs to be booked well in advance of your event, as there are so few in the UK and Europe.
Booking through us is hassle-free and gives you access to a large selection of high-quality verbatim reporters. If you think you would benefit from an on-site speech-to-text reporter, contact us.
You might remember how a court reporter will use a shorthand machine in the courtroom. The court reporter is also known as a speech to text reporter (STTR) or a palantypist, who writes down everything that is heard. Many STTRs started their careers as court reporters – these communication support professionals specialise in working for deaf people.
The short hand machine enables the STTR to write at the speed of someone speaking, at over 180 words per minute with a minimum 98% accuracy. This correlates to one error in four pages. Some very experienced STTRs can write up to 300 words per minute. The STTR can give extra information such as [phone ringing] or [applause], to keep the deaf reader informed of what is happening. The font type, size and colour, and background colour, can be changed to suit the viewer.
A palantypist keyboard
A steno machine
Several keys are pressed at once to generate whole words and phrases. The short hand machine or keyboard is linked to a laptop which has software that converts the shorthand syllables into a written language such as English, so that someone can read the words from the laptop screen. The text can be projected onto a larger screen for a wider audience, or relayed over the internet.
The palantypist adds words to a dictionary in their software, so they build up expertise in the areas they are working in. One of our STTRs has worked for many years in the shipping industry, with all the specialised vocabulary stored in her laptop. Having this pre-programmed vocabulary helps the communication support professional to match syllable clusters to written forms, and write a word much faster. Errors can occur when a STTR mishears a word or where the meaning requires clarification by what is said next. Speech to text is also used for subtitling and closed captions in television broadcasts or webcasts.
US steno keys
Hebrew steno keys
In the UK there are two types of keyboard used by a speech to text reporter; palantype and stenotype. Palantype is British, stenotype is American. The output text appears on the screen in a slightly different way. Different keyboards are used in other countries, such as the veyboard in Europe. The software also varies from country to country.
The British term for a speech to text reporter is palantypist. The American term for a speech to text reporter is ‘CART’ – Communication Access Real-time Translation. CART is also called open-captioning, real-time stenography, or real-time captioning.
In the UK, many STTRs are members of NRCPD, the National Registers of Communication Professionals working with Deaf and Deafblind People. This means they are deemed to have reached the minimum standard of 180 words per minute. NRCPD registration also means they have a duty to safeguard their client, and the client has an avenue of redress in the event of a complaint.
In the USA, CART writers are members of NCRA, National Court Reporters Association. The NCRA has seven levels of qualifications;
Registered Professional Reporter (RPR) Written knowledge test, Literary at 180 wpm, Jury Charge at 200 wpm, and Testimony/Q&A at 225 wpm. After dictation, 75 minutes is given to transcribe notes from each leg with 95% accuracy.
Registered Merit Reporter (RMR) For advanced-level court reporters. Written knowledge test, Literary at 200 wpm, Jury Charge at 240 wpm, and Testimony/Q&A at 260 wpm. After dictation, 75 minutes is given to transcribe notes from each leg with 95% accuracy.
Registered Diplomate Reporter (RDR) Written knowledge test. The RDR is the highest level of certification available to court reporters.
Certified Realtime Captioner (CRC) Captioning training to transition to the realtime captioning field. Written exam and Literary at 180wpm.
Certified Realtime Reporter (CRR) Two-voice testimony at 200 wpm at 96% accuracy.
Certified Broadcast Captioner (CBC) Written test and ability to produce an accurate, simultaneous translation and display of broadcasts using realtime translation software of Literary matter at 180 wpm.
Certified CART Provider (CCP) Written test and five minutes Literary at 96% accuracy 180 wpm.
Speech to text technology has moved into the 21st century. A speech to text reporter can be anywhere in the world, and send live text or captions over the internet to the viewer. This is called “remote speech to text” or “remote captions” when streamed on top of an image. Realtime text is streamed to your PC, laptop, iPad, tablet, iPhone, Android, Blackberry, e-reader, or Google Glass. Your meetings and events can be captioned for you where ever you are.
To book remote speech to text, simply email us. We will email you a link which you click to open a new window for the captions. Your link can be private and password protected, or open to the public. You can ask for your live captions to be streamed on top of your Powerpoint presentation, YouTube video, or on your webpage.
Our team love remote speech to text because it means there is no speech to text reporter in the room with us – it’s just us and our colleagues. You can book a remote writer any time of the day, and at short notice. All you need is an internet connection and a microphone to give sound to the writer. This service can be booked for a minimum time of one hour, then in 15 minute blocks. Our writers have over 20 years of international experience, can write in over 16 languages, and they can understand deaf voices, so they are able to write what everybody says. Your transcript is automatically sent to you when your event ends.
To find out more about our expert speech to text services for deaf people, contact us.
Do you think subtitles on TV have gotten worse? Say hello to respeaking.
A respeaker (or voice writer) uses a mask or speech silencer to repeat what they hear into their computer, which uses voice recognition software to translate speech into text.
Image: www.gccra.org
The people providing live subtitling for television are called captioners. If they use voice recognition they are respeakers. In the UK, respeaking is only used in live captioning, it’s not usually used with deaf people.
Speech-to-Text Reporters (STTRs) are called CART Providers in the USA, or stenographers. They use verbatim shorthand machines which have been around for over 20 years in the UK so they have always been a profession in their own right. The spoken language is not modified in any way, what appears on the screen is exactly what is spoken and this relies on the shorthand skills of the Verbatim Speech-to-Text Reporter.
Respeaking is relatively new in the UK and is seen as something completely different due to the very different skills involved.
The qualification exam to become an STTR is only open to people able to write verbatim shorthand at 180 wpm. In the US, certification is at 225 wpm.
Respeaking is a very different skill: you need to talk fast and very accurately, and know a lot about speech recognition software. The most training time is spent on
The BBC used to pay stenographers to write the live subtitles for TV programmes. To cut costs, they started using respeakers. What is the difference between a stenographer and a respeaker? What are the issues a consumer should be thinking about when deciding which to use? We spoke to Voice4Me*, a large provider of respeakers, to find out more.
A stenographer can write up to 350 words a minute and speed can be an issue for respeakers.
Respeakers are advised to only do 15-minute stints for up to two hours of work as it can damage vocal chords to do any more. How can their voice maintain quality? This has been researched extensively by Voice4Me who have been training and providing respeakers for many years.
Voice4Me only has four out of a team of over 20 who can get 98.5% on regular political TV programmes. They wouldn’t get that on a programme that would be more ad hoc and without much prep. It is not possible to respeak constantly over a two hour period or a whole day, and maintain voice quality. Voice4Me found that the quality drops as the stints get longer. They may claim to be better and to do everything, but the evidence isn’t there.
Not everyone can walk off the street and be a good respeaker. Voice4Me have stretched the boundaries and still find the service wanting. They have had varying degrees of success with their staff, who are very familiar with the industry and the requirements of a deaf audience. Their respeakers are highly trained and are only required to make 97% accuracy. On rolling news, they don’t give them a content/contextual percentage as this accuracy level would drop again. Stenographers are required to hit 99% accuracy and 98% content. Only a handful of respeakers can hit 98% and aways on familiar content, and very very few are verbatim.
What is the difference between 97% and 99% accuracy? It is the difference between stenographic captions and TV subtitles. It’s a big difference. It doesn’t sound a lot but it is. Respeakers don’t all hit the target of 97%; they are supposed to. That isn’t contextual accuracy. That’s just what they get down – the content. STTRs have a contextual accuracy requirement so they have to get everything down – and hit the accuracy level of 99% as well. Voice4Me recognise the differences in skills and their respeakers are better trained than any other. And that is only 15 minutes worth.
In the UK and the USA, stenographers have to be able to write verbatim. This is not the case in Europe. A stenographer will include information such as laughter, applause, and colloquies. In the UK, speech to text reporters are required to have a minimum of 180 words per minute (wpm) with a syllabic density of 2.4. In reality, they have to write a lot faster than that to be verbatim, generally over 200wpm. 180wpm is the entry level.
Many stenographer clients would not accept the accuracy levels of respeakers. Clients often want the verbatim nature of speech-to-text. The danger is you are paying for a service that won’t meet your needs, so why pay for it at all?
Preparation prior to a job is very important and affects accuracy. Preparation time and material for respeakers is much longer and more important than for stenographers. Most respeaking jobs don’t provide either.
There are some great respeakers doing subtitles but they have been specially trained and have been working in the same environment every day for years – not the varied situations stenographers find themselves in. There is a long way to go and respeakers are not the next new thing. They won’t be able to handle the long hours or full days that stenographers do, nor the varied content.
In the UK, respeakers cannot work onsite, they only work remotely. To work onsite, they would need to work in an interpreter’s booth and the audio is piped in, which is not generally available in the UK. Onsite, they can use a special mask to speak into, which softens their voice, however this can still be disruptive. Clients want a discreet service, which will move with them from meeting to meeting, with no fuss.
A stenographer is more cost effective as they can work on their own, whereas two respeakers are required to do the same job, so it costs more for the consumer. Voice4Me’s respeakers do other subtitling work, which is where the value is found in the employee. Respeakers can’t do the volume of output that a stenographer can, and that is what separates them. You need two respeakers to do what a stenographer does, which makes a stenographer cheaper.
There are a few really good respeakers out there but there are many that are not so good. They are also mostly experienced subtitlers so they know and understand the material they are doing really well – such as the news on television. The sound they receive is excellent quality and even the shows they don’t have prep for, they know them well, and will have worked on them before. Voice4Me invested heavily in respeaking, with varied success. They had high quality subtitlers at 99% and now have settled at 97% for respeakers. They have never had so many complaints, and are under huge pressure from deaf groups. People want quality.
Membership of a national register is important because service providers would sign up to a Code of Practice that includes a code of conduct such as;
You shall do no harm, and you will not bring your profession into disrepute.
All registered Communication Professionals in the UK are required to provide an enhanced disclosure and barring service (DBS) check and evidence of valid personal indemnity insurance. This will cover them for a complaints procedure, which protects both themselves, the person who books them, and the client. They are required to carry out Continuing Professional Development on an annual basis as a condition of their registration.
In contrast, with respeaking, there is no real quality benchmark. Professor Romero Fresco’s EU standards are only reliable in the Spanish language. There is no guarantee of quality with the respeaker that you get, no guarantee of personal indemnity insurance, no CPD, no redress for complaints.
To train as a STTR or CART writer, a person would first train as a court reporter. A court reporter using a steno machine could train to be a respeaker in 6 months. Anyone training as a respeaker can learn to use the basic equipment (instead of the newest technologies) in 6 – 8 months. Full training can be completed in one year. Learning the process is not sufficient to reach a working standard; the respeaker must practice and become proficient as well, which takes around 6 months.
The challenges of remote work are usually sound quality and lack of preparation material. Consumers tend to give you fewer breaks as they can’t see the writer. All of these things don’t work well for respeaking. The health and safety aspect was extensively researched by Voice4Me, which means anything more than 15 minutes of respeaking solid text without a break will result in damage to the voice, and also a drop in quality.
Voice4Me has invested a huge amount of money and effort into respeaking. They have in hindsight agreed they should have invested in stenography training but they have invested so much money already, and everything is built around it, so they continue. Their respeakers do lots of other things, not just respeaking, so they are multi skilled and therefore cost effective. Voice4Me has recently taken back on several stenographers as freelancers – who could work day and night for them if they wanted to.
Respeaking works well in captioning because of the nature of the work. The stenographer can work remotely and it’s everything and anything with no prep, bad sound, people mumbling etc etc. Respeakers will be talking and listening, with no breaks. Respeaking is not as skilled a job as speech-to-text or captioning. Like remote speech-to-text, respeaking doesn’t work for every situation and for everyone. Respeakers do not earn as much as stenographers because there is a recognised difference in skills and productivity. If you adhere to health and safety requirements, you need two to three respeakers to do the same job as one stenographer, so are you really saving money? The common goal of all systems is to transform speech to text in (near) realtime and with a high quality of service.
Respeaking might be useful for the educational market. They do have a place in the market in certain situations and with co-working, but they certainly are not at a point where they will be a like for like replacement for stenographers. Consumers will have to decide if a cheaper price is more important than quality issues – and that is their choice.
* Names changed to protect privacy
