About 15 years in the past, I found an unbelievable productiveness hack – I can discuss at 1,000,000 phrases an hour, which is considerably quicker than I can clumsily sort. As a journalist and reviewer, what I wanted to extend my productiveness was voice dictation software program.
There was just one possibility for Mac OS on the time – Dragon Dictate (Sadly, now discontinued however you may nonetheless use Dragon Professional on Home windows). After an hour of studying the Dragon scripted textual content right into a microphone, I used to be prepared to start out speaking as a substitute of typing. The outcome was that it considerably diminished the time it took me to write down articles, however what I actually wished was the power to do that on my cellphone.
I typically make notes whereas out reviewing cameras; it helps me to recollect any key factors that I come throughout whereas out capturing. At first, I might do that on paper, however as smartphones acquired higher, I might use a note-taking app or create a voice recording. Nonetheless, this was completely different from having the ability to communicate totally fashioned sentences and have these spoken phrases saved to a textual content doc. I wished all my fastidiously curated ideas able to be padded out right into a fully-fledged overview. There needed to be a greater, extra environment friendly manner. Roll on just a few years, and there was, and it’s one thing that I’ve been utilizing virtually daily for just a few years.
That is a part of an everyday collection of articles exploring the apps that we could not reside with out. Read them all here.
Otter.ai is a web-based automated transcription service. Through the Otter.ai app, you may hit document, begin talking, and see the phrases that you’ve spoken instantly typed up onto the display. You may, after all, do that with most smartphones for typing out messages and a few textual content, however Otter.ai provides extra than simply real-time transcription, and it has fully modified how I work with textual content.
There may be one other a part of what I do, which includes interviewing photographers to write down function articles. I document with two voice recorders – one from Sony and the opposite from Olympus. I’ve a splitter cable that enables them each to document from an in-ear single microphone that information my cellphone as I maintain it as much as my ear. The rationale for 2 recordings is redundancy, simply in case a recorder fails or a file will get corrupted.
To transcribe these interviews, I used to pay attention again by means of headphones to hour-long conversations, repeating each phrase verbatim into Dragon Dictate on my Macbook to create a transcript. It was higher than listening at half-speed and typing away, however there nonetheless needed to be a greater course of. Once more, Otter.ai got here to the rescue because of the power to add audio information for transcription.
Saving hours of time
Now, when I’ve completed an interview, I plug in one of many voice recorders by way of USB, add the mp3 file to Otter.ai, and go and make a cup of espresso. Because the kettle boils, the water pours, and I return to my desk, Otter is busy turning audio to textual content, understanding who’s talking, tagging every paragraph with a speaker, and doing another cool AI stuff.
Espresso now in hand, I can then sit down and skim the entire transcription on my laptop computer or cellphone and get to work enhancing away. From ending a cellphone name to sitting, enhancing the transcription can take as little as a couple of minutes, and the accuracy is spectacular.
It is not good, however the phrase recognition does properly. Additionally, the app may very well be a little bit extra personalised. I may, as an example, do some digicam and photography-specific phrases. Nonetheless, the transcription is correct sufficient for me to get began. Even when the individual I’ve been interviewing is talking damaged English in a heavy European accent, the transcription remains to be usable. Every speaker is tagged, and will I alter the tag from ‘Speaker 2’ to ‘Dave,’ it’ll undergo and alter each transcript ‘Speaker 2’ reference to ‘Dave’.
Higher nonetheless is that Otter.ai will be linked to your Google calendar, with the cellphone app sending you a notification asking if you wish to open it and begin recording a gathering. Impressively, it integrates with Zoom and Google Meetings to document the audio and transcribe every part that has been mentioned. It may even save screenshots and slides as a part of its abstract, which is the place the AI is available in.
Cool AI Stuff
The app’s AI software program acknowledges matters and might create a ‘Abstract’ of the transcript with timecodes, making it straightforward to rapidly discover the components of the dialog. You may even discuss to Otter in a ChatGPT model to ask it questions in regards to the transcript, similar to ‘write a two-sentence abstract of this dialog.’
Now when I’m out reviewing a digicam, I typically begin Otter.ai and begin speaking; “I’m writing my overview whereas out with the product and providing feedback and ideas as I uncover issues a couple of digicam or no matter I’m testing.”
In fact, there’s nonetheless enhancing to be accomplished, however the work of getting the data out of the mind and right into a digital textual content is finished. I can load the transcript as a .txt or .doc file into Grammarly or Pages and start forming an entire article.
Free, Professional and Enterprise Accounts
Otter.ai is a free service with customers in a position to transcribe 300 month-to-month minutes however restricted to half-hour per dialog. With a free account, you may solely add three audio or video information for transcription.
I’ve a Professional account, which is $16.99 a month or $120 each year. It lets you create groups (every particular person billed individually) to work with transcriptions, and extra importantly, it’ll transcribe as much as 1200 minutes a month, as much as 90 minutes per transcription, and you’ll import ten audio or video information every month.
Lastly, there’s the Enterprise model, which is $35 monthly and $240 each year and permits for 6000 minutes of as much as 4 hours of transcription and limitless audio and video transcription – which can even be nice for YouTubers and podcasters.