Man vs. Machine: Our typists take on Speech Recognition Software
Cassandra CarseWe’re just starting out a new decade and technology today has certainly brought us into the future. Self-driving cars, Virtual Reality, and Artificial Intelligence are no longer a dream, they’re a part of our world.
So, surely, I can transcribe my interviews with a simple voice-to-text program?
Unfortunately, not quite yet.
Keep reading to learn about the history of speech recognition, and its limitations and be a part of the first-ever Pacific Transcription NZ Type-Battle between man and machine…
The History of Speech Recognition
Speech recognition is a technology that most of us have in our pockets.
“Okay google, what’s the weather like outside?”
But how did we get here? Well, the first software was actually invented back in the 50s and took up the entire laboratory wall. Its name was Audrey, and it could understand numbers zero to nine. Each decade the software became more complex and its functionality improved. In 1990, Dragon developed the first publicly available speech-to-text software and by the 2000s Dragon Naturally Speaking could comprehend approximately 100 words/minute.
However, still to this day, there is one limitation. Speech recognition software can only transcribe one speaker.
There are many brands available to the public now. Dragon still leads the market with software that can cost thousands of dollars. It now has the ability to learn your vocabulary, and your speech pattern and can be tailored to suit various professions. On the other end, a quick google online will pull up a number of free programs. Unsurprisingly, these ones can often be inaccurate, meaning you spend more time editing than you would have if you’d typed the document from scratch.
Learn more about the history here.
Typist vs. Computer
Now, the reason you’ve all come here – the never seen before, first of its kind – Pacific Transcription NZ Type-Battle.
We’re putting our typists to the ultimate test. They’ll be pitted against two free speech-to-text software, to find out once and for all, who the better typist is. So, can our typists beat the machines?
ONE SPEAKER
Windows Speech Recognition software
Windows Speech Recognition software can be used in a number of ways. In this case, we will be using an audio player with speech recognition functionality, to transcribe pre-recorded audio.
To make it easy for you to judge, we recorded one of our staff members reading this article using our top-of-the-line Olympus DS-9500 recorders. We then used Express Scribe to run Windows Speech Recognition on the audio file. The results speak for themselves:
But I had Estonia any decade and technology today-border into the future player if you’re planning colours, naturally it, and unofficial intelligence and military, the parallel world.
However, surely, a contract that my interviews with them for wouldn’t expect the
Unfortunately, not quite yet again by
If you begin to learn about the history of the calculation, implementation and be part of the first ever see the transcription yeovil and back panel between man and machine that would leave
The history of speech recognition
The condition of the technology that medal of having a pockets. “Taking the law, but there were no idea behind?”
The family get here? Well period, the software would actually invented back in the 50 and took up the entire laboratory wall again implement Audrey Gillan can understand number you 09 billion in each decade the floor flat became more complex and functionality improved 1,000,000 in 1990, bagging developed the first publicly available speech to text of the and by the 2000 dragon naturally speaking could comprehend 100 wider than the into however, ~ this day the one limitation period each recognition software KM and that one speaker period
The remaining and available to a public company and dragon still leads the market would software that can cost thousands of dollars per year it had by now have the ability to lubricate the line, and yet know and can be tailored to the benefit of the and the other end, applicable line for a number freedom and again under plainly, the ones can often be inaccurate, the new spend more time anything that would have tackling the
Free online voice-to-text software.
There are quite a few free online speech-to-text generators. Unfortunately, none of them allows pre-recorded audio file uploads, meaning the only way to use them is through direct dictation. Regardless, the results were surprisingly accurate, this particular program only had trouble following punctuation commands.
We’re just starting out a new decade and technology today has certainly brought us into the future Point elf-driving cars, virtual reality, and artificial intelligence are no longer a dream, they’re a part of our world Point
I can transcribe my interviews with a simple voice to text program? new line unfortunately, not quite yet Point new line to learn about the history of speech recognition, it’s limitations and be part of the first ever Pan Pacific transcription New Zealand tight battle between man and machine new line the history of speech recognition is it technology that most of us have in our pockets Vine OK Google, what’s the weather like outside?
How did we get here? well, the first Subway was actually invented back in the 50s and took up the entire laboratory wall Point its name was Audrey, and it could understand numbers 0 to 9 inch deco the software became more complex and it’s functionality improved point in 1990 dragon developed the first publicly available speech-to-text software by the 2000s Dragon Naturally Speaking comprehend approximately 100 words per minute however, still to this day there is one limitation speech recognition software can only transcribe One voice
Although we are genuinely impressed, this software is unlikely to save you much time. We got these results by reading off a printed page, very slowly. If we were to speak naturally, it’s likely the program would have made many more mistakes, meaning more time editing. In saying that, a program such as this one is an excellent option for people who find it difficult to type.
MULTIPLE SPEAKERS
As previously mentioned, voice-to-text software is not built to transcribe multiple speakers. Yet, it is a question we receive often. We thought it would be fun to add a second speaker and see just how well the speech recognition software can cope…
Windows Speech Recognition software
We had two of our staff conduct a short interview and then used the same process of running the audio through Express Scribe to utilize Windows Speech Recognition. If you can understand it, please, let us know what it is talking about!
As far as the things that only men who headed a service to tell you where venture after she had left him that he is leaving the eastern coast of throw out the area had 11 U.S. future classroom: and you have tightened documents in one lawyer accused the a nice. The length of the entire city actually is which is lasting effect that has been going through the mail is the best of Mrs. Activists say in its last lead listening at the time of those elected last call seven million transaction and Jason Kidd that was the machine at the investor town will be at passes that is the is roughly flat last minute where is that sometimes accused fatten cattle are a little more than a month ago and she is the main thing is it’s unnecessary as assigning businesses and individuals as a national that passive investors so when we started out as voyage of his DB: does I know it’s just a car and she has a certain people valuable time will be the holiday visit the statement that a key
Our Pacific Transcription Typists
We gave one of our typists the same audio and asked them to transcribe it using the standard Pacific Transcription style. This is what they produced:
START OF TRANSCRIPT
Facilitator: Hi Cassie, thanks for joining me.
Interviewee: No problem, happy to help out.
Facilitator: Let’s jump right in. How long have you worked for Pacific Transcription?
Interviewee: I’ve worked here about two years.
Facilitator: Were you familiar with the industry before starting this job?
Interviewee: I was a little familiar. I had a course in university which required us to record an interview and then type it into a document. Similar to what we would produce here. From that experience I learnt how difficult and time-consuming it actually is, which is why I have so much respect for our typists.
Facilitator: What is your role here?
Interviewee: My role is pretty versatile. Most of the time my job is to check and edit the typists’ work before sending it out to the client. I’ve also worked in the products department so I’m familiar with transcription and dictation equipment too.
Facilitator: What’s one interesting fact you learned from this job?
Interviewee: Well the average person’s typing speed is roughly 40 words per minute. Whereas professional transcriptionists’ type at nearly double that rate.
Facilitator: What’s surprised you most about transcription as a service?
Interviewee: I think just how necessary it is for so many businesses and even individuals. I guess I’ve never given it much thought but outsourcing transcription is so important for researchers, doctors, lawyers, investigators. It’s used in court, to type novels; it’s really just so far reaching and saves so many people valuable time.
Facilitator: Well I think we’ve covered everything. Thanks again for speaking with me.
Interviewee: No, thank you.
END OF TRANSCRIPT
So, man or machine – who won this round?
