I’m rather keen on apps and applications that take my speech and convert it to text for me. I love the time-saving, the convenience and the fact that I don’t have to put more stress on my hands when I am writing a lot. Most recently I have been testing d2u. d2u is an iPhone app combined with a professional transcription service. While in a different league than Dragon Dictation and Vlingo due to its price per minute, it is also in a differently league due to the remarkable accuracy.
To review the app and service I have been using the transcription credits Bottle Rocket, the company behind d2u Transcriber, provided me to actually write this post. Everything that follows and is in italics is transcribed text from d2u that has not been edited or altered. (Translation: if there are errors they were left in intentionally… this time around.) 🙂
This is the first time that I’m trying this particular transcription application on my iPhone 4. I have used a number of voice-to-text applications in the past including DragonDictation and Jott but this one is rather different on a number of levels. Dragon Dictation is currently free. It transcribes what you say using only a computer and it gives back to you within a few seconds. The downside of using Dragon Dictation is that you can only record 60 seconds of voice note at a time and then it has to be transcribed. Yes, you can add additional snippets one after the other but you can’t speak for an extended period of time the way I am with this application and have it transcribed. That noted I’m only now coming to the one-minute point so I could have used Dragon for everything up to a few seconds ago. Jott is a little bit different further. It allows you to transcribe your voice in15- or 30-second pieces (depending upon the level of service that you purchase) and it has to spend some time processing your note before it returns the text to you because it uses a combination of computers and people in order to do transcriptions. The results are transcriptions that are far more accurate than you will find with Dragon if you are not being careful. But it does take 15- to 30-second snippets only and you do have to wait for the text to be returned to you. The advantage of using Jott is that it can integrate with a number of other services such as your Google Calendar or your task list if you use specific services. You can simply speak what you want to add and it will do the rest. So, it isn’t simply a voice-to-text transcription service but it is something far more than that. Worth noting is yet a third application of used which is Recall. It is similar to Jott but allows you 30-second snippets and is far less expensive. And it is a complex task and reminder system not just a service intended for voice to text.
That brings us to this particular application. It allows you to speak for an extended period of time – in fact you can speak as long as you would like to speak. So, right now I am only hitting the three minute mark but I can record for a far greater period of time if I want to. The text, when I’m done with it, will be sent to be processed by the servers and I assume the people who are part of the transcription process which will yield a far more accurate transcription than most of the other services but there is a turn-around period.
For example, my understanding is that if your text is under five minutes, you will get it back within an hour – that means you have to wait at least an hour to get your text back. This isn’t a service for a quick turn around but rather it’s a service for doing accurate transcriptions of longer pieces of texts that you might want to record or dictate such as those that a doctor or lawyer would want to record over time. There is yet another difference with this application compared to the others, that is the price. It cost over $2 per minute once you’re done with your early trial. So this recording so far would already would be costing a dollar. This isn’t something a casual user is what going to want but it is something that a professional who builds in transcription to their expenses may find incredibly useful.One of the things I am curious to see is how accurate the transcription that is returned to me will be and we will see that in just a little while. At this point, however, I am stopping the process of recording because I have to go into some meetings.
I will actually send this piece of text to be transcribed. I will see how quickly it’s returned and what the accuracy is and then I will finish this post and review on it since I am committed to only using the service in order to write this post. By the way, this has just hit the five minute mark now.
One of the things that impressed me immediately upon beginning to use this application and service was the degree of customer service they provide. The first dictation session that I did went about five minutes and three seconds long. The company has a procedure whereby any recording that is five minutes or less from an iPhone will be transcribed and returned within the hour but anything longer than that requires a one-day turn around time. That is certainly not unreasonable. Well my first recording went three seconds over the five minutes. As a result I immediately got notified that it would take a full day for the transcription to occur. A few minutes later, I received a second email from customer service asking me to disregard the prior email. They had adjusted the time down to five minutes so that the transcription would be returned within the one-hour timeframe. That is incredible customer service and a terrific eye to detail that makes a huge difference when you are talking about a service-oriented product such as this. The next thing that I noticed was that when the transcription was returned to me it was perfect.Seriously, it was spot on. No, I was not using any big or complicated words but even the best voice-to-text transcription software I have is not as good as what I got back here.That is an important thing to keep in mind when comparing this product to other, less expensive, products.
It also raised for me the question of how this service would handle technical terms as compared to other voice-to-text transcription products. So in a few moments I am going to do an experiment whereby I am going to read something from a medical website into this application as well as into Dragon dictation and put both up on the site so you can compare for yourselves which does better.It is also worth noting that the iPhone application is excellent.
The interface is easy to use, the buttons are clear and are marked, and it gives you a tremendous number of options with regard to what you want to do with your recording. It has scrubbers so that you can advance or return in a text or in a voicemail more accurately and it allows you to give whatever name you want to the file. You can also choose whether or not you want to upload it to the service, save locally or email it to yourself. Again, all good things to keep in mind when comparing the price of this to other possibilities that are currently available. On with the test…
In order to test out this application and transcription service and it’s accuracy as compared to something like the free Dragon Dictation for the iPhone. I decided that I would read something that had a few medical terms. So, I reached out and found the introduction to Grey’s Anatomy and I’m going to read in to both this application and Dragon Dictation and put up both transcriptions without changing anything. Here is the introduction…
From Dragon Dictation:
(Note I spoke slowly and clearly and it took two 1 minute sessions to read this through.)
The term human anatomy comprises a kind consideration of the various structures which make up the human organism. In the restricted sense it deals merely with the parts which form the fully developed individual and which can be rendered evident to the naked eye by various methods of dissection. Regarding from such a standpoint it may be studied by two methods: (one) the various structures may be separately considered — systemic anatomy; or (two) the organs and tissues may be studied in relation to one another — topographical or regional anatomy. It is, however, of much advantage to add to the facts ascertained by by naked eye dissection those obtained by the use of the microscope. This introduces two fields of investigation, the study of my new structure of the various component parts of the body — astrology — in a study of the human organism it’s immature condition, i.e., the various stages of its intrauterine development from the fertilized ovum up to the. What it assumes an independent existence — embryology. Owning to the difficulty of obtaining material illustrating all the stages of this early development,/be filled up my observations of the development of lower forms — comparative embryology, or by a consideration of adult forms in the line of human ancestry — comparative anatomy. The direct application of the facts of human anatomy to the various pathological conditions which may occur constitutes the subject of applied anatomy. Finally, the appreciation of structures on or immediately underlying the surface of the body is frequently made by the subject of special study — surface anatomy.
The term human anatomy comprises a consideration of the various structures which make up the human organism. In a restricted sense it deals merely with the parts which form the fully developed individual and which can be rendered evident to the naked eye by various methods of dissection. Regarded from such a standpoint it may be studied by two methods: (1) the various structures may be separately considered—systematic anatomy; or (2) the organs and tissues may be studied in relation to one another—topographical or regional anatomy.
It is, however, of much advantage to add to the facts ascertained by naked-eye dissection those obtained by the use of the microscope. This introduces two fields of investigation, the study of the minute structure of the various component parts of the body—histology—and the study of the human organism in its immature condition, i.e., the various stages of its intrauterine development from the fertilized ovum up to the period when it assumes an independent existence—embryology. Owing to the difficulty of obtaining material illustrating all the stages of this early development, gaps must be filled up by observations on the development of lower forms—comparative embryology, or by a consideration of adult forms in the line of human ancestry—comparative anatomy. The direct application of the facts of human anatomy to the various pathological conditions which may occur constitutes the subject of applied anatomy. Finally, the appreciation of structures on or immediately underlying the surface of the body is frequently made the subject of special study—surface anatomy.
I just completed reading the text from the introduction of Grey’s Anatomy into D2U and I have two minutes left in order to reach the five-minute mark that will get this text back to me within an hour. Since I want to get this post up today, I wanted to do this last step and keep it under five minutes so that I get the text back.
One of the things that I will note is that when reading the text into this application, I allowed myself to speak far more quickly because I know that there is a human being during the transcription process and that they will be able to pick up more accurately what I said if I’m faster than an automatic computerised system probably would be.
One of the things that bothered me in this process was the fact that someone at the company knew that I had sent a transcription. They emailed me and asked when the review would be up because they “noticed that I sent a transcription the day before.” The fact is this reveals the fact that there is to some extent for anybody truly concern about privacy some degree of issue involved here. I do not believe they read the transcription text once it’s done and they send it back encrypted but even the fact that they knew that I was sending a piece of transcription would have some people concerned. I’m not but you might be.
So, what do I like about this? It works well, it’s accurate, and if it’s under five minutes you get it back within the hour. What needs improvement? It is more expensive than other transcription applications but it does let you speak for a far longer time and with greater accuracy. If you are a professional and you use a transcription service, this may be worth a look.
The bottom line for me is this – this is a terrific product. The iPhone application is excellent and the transcription service that they offer is phenomenal. It is accurate and if the recording is under five minutes it is also fast. Because of its pricing structure however it is more for corporate users than it is for individuals. I can see doctors using this, lawyers using this, or someone who has a significant budget for transcription services already and who wants to go this route instead of the more traditional route.You do have to keep an eye on the time if you want to get the text back within an hour but that is certainly not difficult to do with an iPhone application that is laid out this clearly.
The app is a free download in the iTunes App Store.
Transcription is $2.28 per minute. (This post would have cost between $30 and $35 to write. In fairness I was speaking slowly which took more time, so figure $25 – $30.)
What I Like: Easy to use iPhone app; transcription quality is superb; recordings under 5 minutes are returned within an hour
What Needs Improvement: Can get expensive quickly, cheaper or free alternatives are available (but not as accurate most of the time and limit the amount of time you can speak); there may be privacy concerns for some