How to Learn General Transcription
Transcription is one of the perpetually in demand jobs on freelance working platforms. These transcription jobs can be found alongside data management, virtual assistant needs, and administrative assistant jobs.
Basically, the role of a transcriptionist is to convert audio to text. Transcriptionists produce an accurate transcription of a recording which can come in any audio or video file type. Plenty of people who want to find easy work on the internet will often apply to be a general transcriptionist, but readying an accurate transcription isn’t as easy as you might think. To be a good transcriptionist, you essentially have to have at least six things:
- Fluency in a widely-used language such as English, Mandarin, or Spanish
- Quick typing skills
- Excellent listening skills
- Thorough research skills
- Technology proficiency
You might think that these traits are common and easy to come by, but you’d be surprised at how challenging transcription actually is. For instance, you might think that if you’re a native English speaker then you’d be a very good English transcriptionist, but it’s not always the case. There are so many accents in the English speaking world alone and not many native speakers are able to decipher each and every one. However, as with anything, practice makes perfect. When you practice often, you learn how to produce excellent transcriptions even with the most challenging factors thrown in like unfamiliar accents, hard-to-decipher audio, difficult topics that are harder to research, or demanding turnaround requirements. Essentially, the more experience you get, the better you become in mastering general transcription.
Where and how can I practice transcription?
First, you would need these tools so you can start transcribing:
- A computer or a laptop - if you want to be a professional transcriptionist, you will need access to a computer or laptop so you can transcribe without disruptions.
- Transcription software - it would be very difficult to transcribe using plain old media players. In order to transcribe efficiently, you will need handy transcription software which should allow you to use hotkeys or pedals for quick replay or transition of audio. There plenty of free-to-use transcription software out there, like Express Scribe.
- Headphones - another thing you absolutely need is a good pair of headphones. You can’t just use your laptop’s speakers because then you wouldn’t hear the audio clearly. You can’t afford to miss any utterance if you’re gunning for an accurate conversion of audio to text.
- Word processing software like Microsoft Word - of course, you also need software like Microsoft Word. While some transcription software may have a built in word processing field, you might need Microsoft Word in order to properly apply formatting. Notepad and WordPad might be enough for some cases, but they can be challenging to work with if you need tabs and styling for labelling speakers and so on.
- Internet connection - last but not least, you will need an internet connection so you can properly research unfamiliar terms or jargon when transcribing. Google is also very helpful when looking for homophones you otherwise might not think of.
When you’ve found yourself complete with these tools on hand, the next step is to find yourself recordings for practice. Here are some ways to practice transcription:
- Go to Youtube and listen to conference speeches or presentations. TEDTalks or Apple showcases are very good kickoff points since it’s usually just one speaker and they have very clear and crisp audio. They also talk about very general and familiar topics which should require minimal research. You can download these kinds of recordings and transcribe them using the transcription software you have.
- Start with short recordings and log your time. Most of the time, popular speeches from TEDTalks should have transcripts you can then use to check your work.
- When you’ve gotten used to one-speaker short recordings, start working with files that are longer with more speakers. For this, you can find free-to-download podcasts or webinars to practice on. They should also have transcripts so you can check your transcription.
- As you progress, find more difficult audio to work with. This can be anything involving speakers with varied accents, or audio that’s not as crystal clear as professionally recorded presentations like TEDTalks. This way, you familiarize yourself with various pronunciations and get used to challenging audio.
Another thing to keep in mind when trying out the transcription route is learning transcription formats. While transcription outfits generally dictate their own thing, here are some of the most common considerations in general transcription:
- Word-for-word or verbatim transcription and non-verbatim transcription
The verbatim transcription format requires you to type everything that is said in the recording. This means including every single spoken thing, including grammatical errors, repetitive sentences or words, filler words and stutters (like ‘um’, ‘ah’, ‘uh’) and even profanity. You might also have to indicate when the speaker does a certain action which can include clapping hands, laughing, and clearing their throat, among many other possibilities. Verbatim transcription is usually needed in legal transcriptions where every utterance from a deposition or legal proceeding is important to be recorded and transcribed for better evidential analysis. Verbatim transcripts may also be required in market research transcriptions – particularly focus group transcriptions – where the accurately transcribed thoughts and ideas of respondents are vital to a market researcher so they can better uncover insights.
On the other hand, when it comes to non-verbatim transcription, you are advised to remove repetitive words, fillers, grammatical errors and other non-essential words or phrases. You can find this transcription format commonly published on blogs or websites, usually as a supplementary to podcasts or videos online.
Labelling of speakers in transcription is a pretty commonplace practice used by general transcription outfits. This can be done in many ways – depending on the specifications for the transcript, you can either label speakers by their name, by their title, or by using common labels like Interviewer/Respondent or Male/Female. Some transcription services may use identifiers like F1, F2, or F3 to distinguish three female speakers for instance.
Timestamps refer to markers of specific recording intervals. It allows the reader of the transcript to easily identify which part of the recording corresponds to which segment of the transcription. Although you can’t see them while watching a captioned or subtitled video, timestamps are very important behind the scenes. In subtitling or captioning, timestamps are invaluable in properly placing the segment of text to the appropriate part of the video. The format of timestamps in transcription may differ depending on their use. For example, subtitling requires timestamps down to the millisecond, in the format [HH:MM:SS.MS]. But simpler timestamps for reference’s sake can usually just be [HH:MM:SS].
- Using markers or tagging with audio descriptions
Aside from timestamps, there are different, more simple markers that are also commonly used. How these markers are transcribed are also highly preferential like speaker labels. Some people prefer very little markers using only the most common ones, like [Laughter] or [Background Noise]. However, in cases where transcriptions may be used in subtitling or video captioning, it is necessary to include contextual descriptions like [Side Conversation] or [Shouting] so that the whole situation is understood by describing a bit of context surrounding it.
There you have it! If you take note of what we’ve outlined here, you should be able to start your general transcription career armed with the right tools and a familiarity of the formats. The key here is to keep working on it and you will become an expert in no time.