This is a very topical question especially since the release of the latest iPhone 4S with its built in personal assistant Siri which has firmly thrust voice recognition technology in the face of the masses.

The answer has to eventually be yes but in my opinion not for a very long time for all types of audio/video transcription. Already we are seeing single speaker audio move away from a traditional typist and into the domain of Dragon – NaturallySpeaking on Windows, Dictate on Mac. Where Dragon currently falls flat and where traditional outsourced transcription services still add great value is in multi speaker audio and video like meetings, interviews, focus groups, podcasts etc.

Our sister company Dictate Australia recently wrote an extensive blog post on using Dragon for multi speaker audio and the transcription options open to you if you don’t use Dragon to help you, titled Can Dragon NaturallySpeaking Transcribe Meetings or Interviews?

Like every business transcription services have to adapt with the times. At The Transcription People we are doing just that. Having recognised years ago that Dragon was going to be substantial in the digital voice world we set the team at our partner company Dictate Australia the task of knowing Dragon, for Mac and Windows, inside out, to allow us to not only provide that software to our clients should they decide to go the voice recognition route but also to act as consultants allowing us to advise on all aspects of speech recognition technology.

There is still a major role that transcription services can provide in the speech recognition space by acting as a conduit for busy users to process their digital audio recordings into Dragon for them. With Dragon NaturallySpeaking Professional edition clients only need to spend an initial five minutes training their voice profile. Once done they can then pass that profile to us. All the client then needs to do is record their digital dictation and send that to us for processing. Our proofreading team then step in and provide a voice recognition service which not only passes the audio through Dragon NaturallySpeaking using the clients own voice profile but they add further value by proofreading the completed document within Dragon thereby updating the user profile and improving accuracy after each transcription.

The client therefore has a hassle free transcription solution using Dragon NaturallySpeaking, all they need to worry about is recording the audio – we take care of the rest. This is a much cheaper solution than traditional transcription as the voice to text is handled by Dragon, the value add comes with the proof reading and profile training.

Quality is everything … isn’t that how the saying goes? Well, it does if the quality pertains to audio for the purposes of transcription!

Too many times we receive audio files for transcription where the quality doesn’t come close to being of excellent quality let alone audible, like it could or should with today’s technology and extensive range of digital voice recorders. Unfortunately, when we advise a client that due to the audio quality, we will only be able to achieve 50% of the dialogue, they ask us to “go ahead”. Little do they realise it can sometimes be just a setting on the recorder or the wrong recorder for the recording setting or not being aware of background noise when recording not to mention blowing out the transcription time which is costly for them. Simple things, yet vastly important, when capturing excellent audio quality for transcription. And to be honest, there’s nothing worse than a transcriptionist having to sit through badly recorded audio, not to mention damaging to the inner ear drums!

The good news is an excellent quality digital voice recorder is not expensive so minimal outlay for more audible audio and therefore a cleaner transcript.

We recommend the following digital voice recorders:

Two speaker meeting/interview:

Olympus WS-750M - small notetaker with 4Gb of on-board memory which can be expanded using the microSD slot, not that you would need to, 4Gb will give you 32 hours of continuous recording on the highest quality MP3 recording mode. Powered by a single rechargeable AAA battery (or with a standard alkaline AAA) this tiny recorder will give you superb audio quality for your one on ones.

With its huge amount of memory (4Gb) which can be expanded even further if you are one for utilising flash memory via the microSD card slot this little recorder also doubles as a handy memory stick to keep your music (which you can play through the recorded and listen through earbuds) or any other digital files you might like to keep handy. The recorder itself is very small and lightweight and will pick up great audio even when tucked into a shirt pocket. It weighs a tiny 51 grams even with the battery in and is a smidge under 10cm tall, 4cm wide and a fraction over 1cm thick.

Focus groups, larger meetings:

Bigger groups calls for a beefier recorder, ladies and gentlemen we proudly present the Olympus DM-3. New to Australia, this recorder only arrived in July 2011 and took over from the very popular DM-450 recorder. Housed in an aluminium shell this is a sturdy recorder with a full colour LCD screen and superb audio pickup. Also a 4Gb recorder and powered by a rechargeable battery OR via USB power from your PC or Mac this unit will give you  just over 24 hours of continuous recording in its 320kbps high quality MP3 recording mode which is perfect for large meetings. You also have the option of stepping up to PCM recording (higher than CD quality) uncompressed WAV audio in which case you can squeeze just over 4 hours recording in this mode.

Both recorders are Mac and Windows compatible, record in either mp3 or wma audio formats and need no software to access the audio files, simply USB plug and play. Both are also extremely easy to use, power on, hit record and thats it.

When using a transcription service one thing you must check is the security of the file transfer mechanism used by that transcription service. The vast majority of audio we process at The Transcription People is confidential, this means that the audio our clients send to us must be sent to us across the internet securely, this means encrypting the audio from the sender all the way through to our servers.

How do we encrypt audio and what should you look for when hiring a transcription service?

There are a couple of common ways to transfer files across the internet, email and FTP (File Transfer Protocol). Neither of those are secure: email by default travels in the clear across the internet from the sender to the recipient. By “travels across the internet” we mean your email is passed from server to server, called hops, until it reaches its destination mail server. As email is sent in the clear the content could potentially be intercepted and read by others. FTP is also a very common way to move larger files and many smaller transcription services make use of this. FTP is fine for non confidential audio, using FTP again transmits data across the internet in the clear, unencrypted. When using FTP you are usually required to enter a user name and password to connect to the FTP server, that is just the key to open up access to a location on a server to deposit your file, it does not mean that your audio is encrypted. Both email and FTP can be secured; for email by using secure certificates for both the sender and receiver and for FTP to use the SFTP protocol (Secure File Transfer Protocol) which uses an encryption key exchange as the transfer is initiated, both of these are rarely used by the majority of small transcription services around Australia.

Recently there has been a boom in third party online file transfer services like SendThisFile.com which some transcription services have embraced. This allows senders to deposit large files, like audio, with the third party. This system does allow for secure file transfer but neither the sender or the receiver have any control of where the data is stored and who has access to that data.

Whats does The Transcription People do to secure file transfer?

At The Transcription People we use web browser secure file transfer across HTTPS direct on our own servers. HTTPS is a combination of Hypertext Transfer Protocol across SSL (Secure Socket Layers), which basically means that whatever is transmitted through the browser is encrypted. This is the protocol used by banks and credit card companies to secure data exchange between the browser and the server and is indicated by the padlock in the browser and the URL starting with https. Using this protocol the data is encrypted from the sender to the receivers server, across all internet hops.

We own and administer our own webservers, we do not share them with other websites or businesses and we have full access and control over our servers. Audio you securely transfer to us via our innovative and easy to use web application comes directly to us, securely across  HTTPS.

Olympus have long been the market leaders in both tape and more recently digital voice recorders and dictaphones in Australia and I am sure in most parts of the world. Olympus Voice in Australia is run from their head office in Melbourne with an additional office located in Sydney.

In 2009 Olympus Australia have been very agressive with updates to their product range and we have seen a number of recorders discontinued and replaced with newer models. The greatest impact of changes in product line have affected the digital dictaphone range, most commonly used by for single speaker audio and very popular in the medical and legal fields. The product range now comprises the DS-2400 (replaced the DS-2300), DS-3400 (replaced the DS-3300) and the DS-5000 (replaced the very popular DS-4000).

All the Olympus digital dictaphones now by default record in the new DSS Pro (Digital Speech Standard) audio format which creates  .ds2 audio files. This is great to see Olympus moving forward with product lines which are advancing with audio technology but this can give smaller transcription companies some issues. DSS Pro is a licenced audio format meaning its audio codec is not openly available restricting transcription of these audio files to companies or individuals with official Olympus transcription module software, generally bundled in an official Olympus transcription kit. Many small transcription businesses rely on free or open source software for transcription, the most common free transcription software is Express Scribe from NCH SwiftSound, ExpressScribe is not able to play .ds2 audio files.

The Transcription People use a mix of transcription software, we regularly use ExpressScribe but that is also complemented with some of our Windows PCs and Macs running Olympus DSS Player Pro 5 transcription software for the times when we receive .ds2 audio files from our clients, a trend which is growing as more clients start to upgrade or replace their digital dictaphones.

If you use an Olympus dictaphone and are looking for an Australian based transcription service, one question you should be asking when you shop around is “can you transcibe .ds2 audio files?”.