Day 9 : Podcast Transcription : Unleashing the Full Potential of the Spoken Word

black_microphone_on_baby_blue_background

Today I spent another hour working through the remaining TranscribeMe training modules while simultaneously going through the style guide one page at a time. The training modules provide good information and some practice on the kinds of grammar and proofreading skills you will commonly use when transcribing the audio, in addition to the markup tags which are used to deal with any factors in the audio which are either not formally in the realm of speech transcription proper (ex. [silence] to denote a speaker’s silence for tens seconds or more, [inaudible] to denote that the speech is simply too unclear (at least for you) to decipher, and various other useful ones). Learning to use the tags properly allows you to deal with essentially ANY issue which comes up in the audio which acts as an obstacle to you properly transcribing every word in the project.

When I completed the training modules I simply moved in the same manner through the test. Since it is possible to answer a few questions and then log off of the internet, log back on and simply resume the test, I proceeded to take my time with the exam while also continuing to work through the style guide. During the test itself there were a few instances where I needed to directly refer to the style guide in order to answer a question (or transcribe one of the audio files) properly.

Most of the test was straight forward and the questions could be easily answered correctly by anyone who had paid attention during the training and read carefully through the style guide. For some reason, however, in the second part of the exam (right before the end) I ran into some difficulty in the transcription section which I couldn’t seem to figure out. In fact, they give you a certain number of tries to get it right and then, if not successful, you must restart the whole test. So after trying twice I was getting worried that I would have to do all the work of restarting the exam. Luckily, my research instincts came to my aid. After carefully re-reading the style guide a few more times and still not being able to get past the trouble spot in the exam I decided that I needed to try and find additional information about the TranscribeMe system. Luckily, my instincts were correct and I found the answer in one of the videos on the TranscribeMe training videos YouTube channel. In the video they explain that since the audio files have been broken down into micro-chunks then you can never be fully sure if the first word in the file you are working on is the beginning of the first sentence or a continuation of a sentence or questionfrom the previous chunk. Therefore, you are suppose to NOT add ellipses (…) at either the beginning or the end of the file. Unfortunately, although the style guide is generally quite good it does not make this point clear enough (which is the main reason why I don’t feel guilty for mentioning the issue here). I am also aware from a contact who was doing the exam at the same time that they were getting stuck at the same point in the test. I was also to find out later – once I had access to the internal social media message board – that many people had had this problem at this same point in the exam. Fortunately for me, my research training had saved me from unnecessary hassle. Again, the strategy of taking in as much information about a subject from as many sources as possible enabled me to find a solution to this issue without having to bother anyone or wait for a response. This same research strategy has safeguarded my from various problems in the past. As a result of my diligence I avoided the frustration and waste of time of having to repeat the test. In fact, once I found the solution I gave the test one last try and – with fingers crossed and by making the appropriate changes – this time I was relieved to see that I passed.

Once the exam was passed a page came up which explained that I would be contacted shortly by an administrator who would direct me in how to access the workhub and begin working on audio files, interact with the community, and all of the other good things that certified TranscribeMe transcriptionist are able to do. At this point I simply continued working through the style guide, blog posts and other resources I had collected (of which there were already enough to keep me busy for months). Once the administrators contacted me and gave the green light I would simply add the new resources and tasks which then became available available to the overall mix and “things to do” lists. It is all just one continuous training process.

The Connection Between Transcription and Podcasting

I have always had a serious interest in podcasting. I am constantly listening to various podcasts each day and I find that the format has some unique features which make it better than printed text or video in some circumstances and for certain purposes. For instance, on many days I spend so much time sitting up looking at the computer screen and I just want to be able to lay down and listen to some interesting content. In fact, I often schedule my research to alternate between “sitting up” tasks and “laying down listening” tasks. This is much healthier for the back and eyes than constant screen exposure. You can imagine what great pleasure I have found in the fact that the first transcription company I started working for within the last few weeks issues a continuous and eclectic supply of podcasts for transcription and editing.

Here is an excellent video produced by Cliff Ravenscraft, a well-known podcasting expert and consultant, where he discusses four of the top unique benefits of the podcasting format over other forms of modern communication :

Two of the unique benefits which Mr. Ravenscraft’s discusses and which I agree are most interesting are

(1) the effect of the increase in smartphone usage to provide a rapidly growing audience for podcasts and

(2) elimination of mandatory “screen time” (looking at the computer screen) to consuming podcasting content

These are two important factors which I believe will only increase the popularity and consumption of podcasting into the future. For instance, think about all of the people who are willing to listen to a podcast while driving to or from work (especially in traffic), or while doing chores, or as a break from sitting up looking at the screen.

The people over at TranscribeMe have devoted a whole blog post to the issue of podcasting and its relation to transcription : Podcast Transcription where they point out the essential benefit of podcast transcription lying in making the podcast more possible to be indexed in the search engines, since search engines only index text – not audio. They don’t index video either and so the same benefits of transcription apply to video as well as explained in the article Reach More Clients by Transcribing Your YouTube Videos.

Here is an example of a reputable podcast production which includes the transcript for each podcast right on the main podcast page : The Paleo Solution – Episode 112. On this page you will see a link for downloading the (audio) podcast itself and a seperate link for downloading the transcript (in pdf format).

It’s important to keep in mind that one of the most beneficial reasons to transcribe a podcast of video is that it then enables disabled people to access you content. For instance, deaf people specifically can consume your content by reading the transcript. Blind people would naturally be able to consume the audio podcasting content unassisted, but for video they would need a computerized reader device to read a transcript of the content. In either case, transcribing helps to include a wider audience of people who can benefit from your production. For some podcasts, this group of people may constitute a significant portion of their target audience.

Another interesting article from Forbes Funnymen and iPhones: Why the Podcast is Finally Coming Into its Own looks at hot the podcast has allowed those who have less mainstream views to enter the broadcasting market whereas they would have little success approaching traditional networks. Some of these avant garde podcasters do indeed end up succeedin either through the growth of their podcast alone or through being picked up by one of the larger mainstream networks. Therefore, podcasting is a sort of experimental medium which enables new ideas to build ground and eventually become successful.

As I mentioned before, I have been focusing on editing the audio files (including podcasts) from the one company I work for as the transcription files are rather long and difficult for me at my current experience level. Editing these files, however, is very productive and enjoyable. Since most of the transcription text which comes in attached to the audio file as part of the editing project is of relatively decent quality (some more than others) the reality is that if the content of the podcast is straightforward (ex. not too technical) and the quality of the audio is decent (which is usually the case, since most serious podcasters who are actually willing to spend money on transcription spend considerable effort trying to get the best quality production that they can) the process of editing basically involves a rather leisurely listen to the audio file in real time and following the text to proofread for mistakes. This is called “proofing to audio” in transcriptionist jargon, by the way. If there are errors they can usually be spotted and fixed quickly by stopping the audio for a few seconds and making the quick adjustment. Common errors which occur include : the transcriber omitting an important word, simple spelling mistakes, simple punctuation mistakes, the wrong word which can often more easily be deciphered by the editor who is approaching the audio with fresh ears, etc. The editing step of the process certainly IS important as just having an additional person (who naturally possesses a unique skills set and often a higher level of experience in order to reach the editing level) to go through the transcript and audio can catch most of the mistakes made by the original transcriber. I could safely say that in EVERY SINGLE transcript I have ever edited there has been AT LEAST one error which I was able to correct. Most of the time there are several. So, I don’t feel guilty for choosing the less stressful editing jobs for now, since my work as editor actually DOES improve the final transcript product in some way.

Ultimately, working as a transcriber puts you into contact with a regular stream of interesting and new podcasts. There have been numerous cases where I received a podcast from the transcription company which I enjoyed so much listening to during the editing process that I went out to locate the podcast, subscribe and become a regular listener myself. This is just another one of the juicy perks of being a freelance transcriptionist.

Day 8 : The Computer-Human Hybridization Movement – Increasing Efficiency While Decreasing Unemployment

A quick Google search this morning produced the YouTube video of an excellent speech by TranscribeMe CEO Alex Dunayev at the Silicone Valley Open Doors Investment Conference in 2013.

Mr Dunayev delivers and well articulated and down-to-earth presentation which clearly details the important trends arising in the transcription industry. Some of the most important of these include : the rapid growth of the transcription market worldwide as a result of greater reach of the internet and mobile networks, how new business models (such as crowd-sourcing) are being made possible by advances in technology which are enabling the delivery of higher quality transcription services to a widening customer base. Mr. Dunayev also explains the integration of evolving speech recognition software and artificial intelligence into the transcription process, which is enabling transcription to be done in a more efficient and less costly manner, and thus provide transcription services to individuals and organizations who simply hadn’t had the budget to afford it in the past (ex. students, educational institutions, freelancers, small businesses. He also gives us a promising glimpse of the future potentials of the new transcription paradigm, including ways it will aid disadvantaged populations such as the disabled (ex. blind and deaf), researchers, creative people, etc. Judging from the response of the experts on the questioning panel, who seemed to be quite impressed, I believe that most people come away from viewing this presentation with an expanded understanding of the topic of transcription, as well as the various additional topics and organizations related to the transcription industry. In addition, it appears easy to gain a more optimistic sense that computer technology actually CAN be harnessed and utilized in a practical and humane manner to solve important real world problems (ex. human, business, academic, etc.) while at the same time being easy to implement and affordable to the general public.

In the past decade, this philosophically fragile issue of the “Rise of the Machines” has grown to apocalyptic proportions for many, as theories range from robots taking over human jobs and making us obsolete, to artificial intelligence being taken over by the computers themselves, who then turn on their human creators and initiate a global cyborg war – perhaps catalyzed by the computerized scanning and transcribing of uncensored human thoughts, leading naturally to World War III between the humans, and with supercomputer controlled neutron bomb attacks resulting in mass extinction of the human population of the planet, and allowing the robots to live in a highly organized and efficient utopia until the end of time.

Most informed and sensible people realize that any technology is merely a tool, and that it is the APPLICATION of that technology which determines its ethical value. Examples now abound of new applications of technology able to solve REAL problems for REAL people which have never been possible in the past. For instance, as supercomputing technology becomes faster and more powerful it is being used to analyze data in the field of medicine to gain better understanding of genetic factors in disease, the nature of epidemics, etc. Smartphone apps are being developed which facilitate a growing range of medical treatment processes, often conducted by the individual in the comfort of their own home. To give a few of examples, there are now operational apps which measure and remind diabetics to check their blood sugar level at scheduled times which are making it much easier to control this chronic disease *and various others). New apps which track disease epidemics are allowing public health officials to better protect human populations from outbreaks, and to eliminate outbreaks when they occur as a result of the ability to obtain data related to the epidemic faster. Stem cell technology is showing the promise of regenerating dysfunctional body parts and even restoring various important neurological functions in the body (ex. vision, movement, paralysis). From these few examples most people would probably agree that there are significant potential benefits to the development of these advanced technologies. The fundamental issue thus becomes ensuring that the applications of this higher technology are directed at solving REAL problems for REAL people, with the main goal of improving the lives of people throughout the world.

Along this line of reasoning, I believe that the transcription related technology, service and employment system which TranscribeMe is developing is an ideal example of the proper use of the emerging advanced technologies. It is also an excellent example of how it is completely feasible to integrate computer technology and human capital to ultimately increase OVERALL benefits for the humans who are served by these technologies. For instance, the TranscribeMe crowd sourcing production platform contributes two main benefits to the operation. It enables greater efficiency and faster turnover of the end product (transcription) to the customer, while at the same time fulfilling more of the needs of the transcribers to be able to work anytime, anywhere and more flexibly as they general work on quick (roughly one minute) segments of speech. In addition, since Mr. Dunayev explains that there are just certain limits to what computers can achieve in regard to processing human speech, we see that the computers have a very valuable role to play in the more logistical and technical aspects of the transcription process (ex. splicing audio files into ~ one minute micro-chunks, distributing the micro-chunks to the most suitable members of the transcriber crowd based on demographic data stored in the system, etc). In other words, the computer is acting in a similar way to the timer on a dishwasher or coffee machine. It COMPLEMENTS the human labor.  The computer performs the more menial tasks of scheduling and organization while the skills of the human are used for those elements of the workflow which are beyond the natural realm of the computers. It is the same case for digital music production. Sure, digital musical instruments can do many amazing things which human musicians generally can’t do on their own (ex. synthesizing sounds which don’t actually exist in nature, optimizing music and sounds after they have been recording via advanced digital audio editing software, etc.). Before these technologies were available to humans they had less creative options to work with sound and produce the amazing works that they can now. However, digitized music will NEVER be able to truly mimic the unique human quality which is brought forth through music.

There are certainly some rather ethically unsettling developments playing out in various pursuits which have a strong effect on humans, all other lifeforms on earth and the environment. Some more down-to-earth examples can include : the detrimental effect of information overload on the human brain, the often intrusive nature of Smartphone technology by which people become addicted and neglect more important issues in their lives, the sedentary lifestyle many people have descended into as a result of the technology making it less necessary to be physically active, etc. While each of the above examples can be partially alleviated through the application of proper behavioral (and other) psychology strategies, the bottom line is that humanity is facing a growing challenge of trying to strike a healthy balance of utilizing these helpful new technologies as opposed to allowing the technologies to exploit THEM.

This is why the kind of technology application which TranscribeMe has developed offers much hope in the sense of being evidence of the ability to design technology in keeping with the ultimate practical needs of the humans who actually use the product in their real lives. The computers are not the end consumers of a product like the TranscribeMe transcription. They are the TOOL which enables the end product to be produced in the best form and in the most efficient manner possible. Ultimately, it is the HUMANS who benefit from the fruits of the technology which TranscribeMe has developed. That is, BOTH the end consumer (who receives a very high quality (accurate) transcript in an increasingly short amount of time and at an increasingly affordable price) AND the worldwide crowd of transcribers who benefit from a decreased risk of under or unemployment, job flexibility, high quality training and career advancement opportunities. In the end, the TranscribeMe system is very people-friendly.

At the SAME time, the TranscribeMe system is also computer-friendly. That is, as explained in the presentation, the artificial intelligence of the transcription software actually learns how to better process a specific (repeat) customer’s projects based on all of the data collected from past projects. Therefore, the computers themselves are also evolving and benefiting through the performance of their intended actions (through the combination of big data processing and artificial intelligence).

In the end, a very positive feedback and production loop arises as the computers’ evolving artificial intelligence increases the productivity of the process and thus completes the transcription for repeat customers FASTER. This then frees up resources to be able to process more customers, which increases revenue, which then enables the company to invest more profits into growth and marketing, hire more transcribers (and other necessary workers) and thus stimulate employment and the general economy. Again, this line of reasoning shows clearly how this system delivers REAL benefit to the lives of REAL humans.

What is even more promising is that, as Mr. Dunayev details in the presentation, TranscribeMe has begun planning and implementing some very powerful collaborative projects with other companies and industries which can benefit from the integration of TranscribMe’s crowd sourced, computer-hybrid transcription technology with the their own applications. One example of this is the collaboration between TranscribeMe and NVivo, one of the leading research platforms for data analysis. A significant component of research of most kinds (ex. marketing research, qualitative social science studies, focus groups, etc.) involves collecting data in a form which is suitable for, and optimized by, quality transcription. Proper transcription of data enhances the ability to organize, manage and analyze data with the end result being better quality research, and maximum application of the output of that research.

This integration of TranscribeMe technology into a growing number of appropriate and related applications is positioning TranscribeMe technology to serve as a powerful tool whose function is to convert audio speech into the most potentially accurate text which can then be imported into other applications which use text data as one of the primary inputs. In line with the old saying, “Garbage in… garbage out.” the TranscribeMe technology is minimizing the amount of garbage going IN, and thus acting as a major force for improving the quality of all research which uses real speech data from any source (audio, video, etc.).

I realize that this post has grown extremely long. When I become interested in a subject the words just spewing out of me and it is better to just go with it. It is a natural tendency. A blessing and a curse of sorts. I’ll admit that I have a “writing problem”, in the sense that I often can’t write fast enough. This is, of course, is worsened by my “drinking problem”, where I can’t drink fast enough. Then again, that all depends on the type of drink (ex. beer, coffee, etc.). The reader is, of course, free to take what they want and leave the rest.

But I digress.

Having said all of this, I do TRULY believe that the issue of the proper integration of technology with human nature is one of the most important of our time. Plus, it only seems to be becoming MORE important, and at an ACCELERATING rate as the evolution of the technology itself is accelerating in a non-linear progression. I also think it is important for anyone who has an interest in, and/or wants to work in, the transcription field to learn about this issue, think about and consider the ways (both positive and negative) in which it effects their everyday lives. When I look at developments like TranscribeMe it makes me very hopeful that we are at CAPABLE of developing ethical collaborative integrated applications which utilize the power of advancing technology with the ultimate purpose of improving the human condition.

Getting back to the transcription training issue, in this post I haven’t yet specifically discussed much related to my progress. I have now passed the TranscribeMe application and started on the training phase before being cleared to work on projects. In a way, however, as I mentioned in the last post that one part of my research as I proceed through the transcription world is to watch videos related to the different topics, companies, industry people, etc. Therefore, this whole blog post essentially describes a valuable part of the research process. That is, the process of becoming more familiar with the transcription company I am now working for, getting to know more about how they operate their business, learning about what the company has planned for growth and development into the future, etc. In the same way that an investor does serious research on the “fundamentals” of a prospective company before making the decision to invest in it, it is similarly important to research a company you intend to WORK for to ensure that the philosophy and goals of the company are in line with yours to an adequate degree. After all, when you work for a company you are dedicating quite a bit of your energy and time into the endeavor. Thus it is essential to do your homework in order to make the most informed and prudent decision based on the specific nature of your situation. In addition, what is so great about living in the “information age” is that there is just so much information available if we know where and how to look. That is one of the functions of this blog, of course. One of the main goals here is to teach you (by example) a productive strategy of navigating through the transcription world (and the worlds connected to it) with the ultimate goal of helping you to make the best decisions possible which will help you achieve your goals and maintain a healthy level of continuous growth and prosperity – on the physical, mental and spiritual levels.

As for the TranscribeMe training, I have been working through the training modules while simultaneously reading through the style guide. I should be done with the training by tomorrow and then will attempt the final exam for the training. If/when I pass the exam the administrators will then contact me within a few days and provide me with my login information so that I can access the system, start becoming familiar with how things work, and spend some time browsing the internal social media network in order to begin networking and connecting with some of my new co-workers, etc. I also have plenty of research content to keep me busy both before and after I gain formal entrance into the system and start working on transcribing some of the one minute (or less) length audio files. My main focus, however, is to proceed slowly, steadily and methodically, in order to take it all in at a healthy pace while also enjoying the process of growth and discovery.

I hope you have gained something valuable from the information in this rather long post. In the next post I will further discuss some of my insights on the training and research processes, and do some more analysis of other interesting aspects of the TranscribeMe operation.

TranscriptJunky@gmail.com
https://twitter.com/TranscriptJunky