Audio Transcription – An Essential for Accessibility

How speech technology will help your company keep track of your next meeting, conference, or business call

Using professional transcriptionists and our Artificial Intelligence-enhanced automated transcription tool (CPSL Transcript) we will provide fast and accurate transcriptions of any audiovisual content you need.

What is Transcription?

Whenever a written version of an audio or video file exists, you can be certain that it is the work of a transcriptionist. Transcription is helping us to transform speech, sound, or sign language into readable text, using transcription software and speech technologies. Transcription makes the source material easier to understand and faster to grasp. As there is always a demand for content to be available to as many people as possible, transcription becomes a tool to aid in spreading this information. Technology helps us to use our time more wisely, dedicating it to added value tasks, while the machines work for us.

There are three main styles of transcriptions, depending on the end purpose for the transcription data:

1. Clean read transcription

This style (also known as “clean verbatim” transcription) focusses on clean and readable text for professional and academic contexts, such as business communication, academic research, seminars, or presentations. Unless stated otherwise, a clean read transcription will contain no details that are irrelevant to the content of the audio file, while still retaining the meaning of the sentence. Therefore, slang, interruptions, false starts, etc. are left out.

Example:

Audio: I, I’m, er [pause] wh, what was the question again [laughing]?
Transcription: What was the question again?

2. Smooth verbatim transcription

For purposes focused on transmitting the actual content of the source material, rather than on cleanliness, the smooth verbatim option is a good fit. The audio is transcribed word-for-word, whilst removing filler and non-words like ‘uhm’, ‘hmm’, ‘uh-huh’, and interviewer responses like ‘okay’ or ‘I see’. This style is ideal for interviews, as it relays what’s being said throughout the recording, without cluttering the text with unnecessary detail.

Example:

Audio: I, I’m, er [pause] wh, what was the question again [laughing]?
Transcription : I, I’m [pause] wh- what was the question again [laughing]?

3. Strict verbatim transcription

If there is a need for a detailed transcription of each word, sound or non-verbal communication, the required style would be strict verbatim. Here every sound is interpreted and transcribed, so as to give an accurate 1:1 account of the contents in the recording. This transcription style is used mainly in a legal context, where the way in which a person says something is as important as what is being said. The transcription will include every utterance, false starts, noise, and the like.

Why is transcription important in our society?

Information is moving at increasingly higher speeds, requiring content to be ever faster to grasp and process. Simultaneously, the average attention span seems to be decreasing. If a video does not grab the viewer’s attention in the first three seconds, it is very likely they will not watch it further.

Subtitling, transcription, translation and voice-over of various types of media, often required in your company Meetings, Conferences, Telephone calls, Recordings, Interviews and Podcasts, help to convey information in the most upfront way possible, so there is no need to turn on the audio: it is the most efficient way to relay information.

Apart from the aspect of clarity, there is another reason why transcription is relevant in our society: accessibility. According to a study by the National Institute on Deafness and Other Communicational Disorders in 2011-2012, 14% of adults aged from 20 to 69 were suffering from hearing loss in the US alone. While not being a critical disability, deafness is preventing those who are affected from accessing information that is presented only in audio form, such as discussions, lectures, or broadcasts that are not transcribed. Thanks to transcription and subtitling, the pool of accessible knowledge, like eLearning for people with hearing-disabilities, is growing and paving the way for a world with better opportunities for them.

Why is transcription important for your business?

As a globalized and inclusive business, you will want to reach as many customers, partners and employees as possible. By transcribing and localizing business media, such as conferences, podcasts, and training videos, you can assure their accessibility and clarity to consumers of any language.

This effectively increases the reach of your business and, in turn, causes your business opportunities to grow around the globe.

Automated Solutions – The future of transcription?

Due to the development of Artificial Intelligence through Machine-Learning, applications are becoming increasingly sophisticated at interpreting and parsing patterns like speech. You can already see the advancements through technologies like automatic subtitling for videos. While exciting, these programs are not yet refined enough to handle every quality of video. Fast speech with multiple speakers talking over each other and with loud background noise and bad video quality makes automatic transcription prone to errors. Closed captions are similarly difficult to generate, since noises cannot be categorized well due to the program being unable to make out a clear pattern.

Professional transcriptionists usually employ manual and automated work in tandem. First, a rough draft is produced by the automated transcription software, which is then manually edited and refined. Our automated transcription tool (CPSL Transcript) uses artificial intelligence for speech-to-text conversion. The resulting automated transcription, which is usually ready in minutes, is sent to a professional transcriptionist for review, depending on the quality standard required by the client (unedited or edited automatic transcription).

This saves time and leads to a good balance between quality* and quantity.

*Regarding to our commitment to ensuring the security, confidentiality, and integrity of your data, we have earned our fourth certification, the ISO 27001 information security standard.

There are several cases (with different types of content) where a trusted language partner like CPSL will be convenient when it comes to delivering accurate/edited transcription output: