Many of us have had to transcribe a recording of some type at one point or another. Maybe you wanted a quote from an important interview. Perhaps you make personal voice memos and then write them all down later. Or you may just want to generate a searchable text from a long speech.
That's where transcription services can be of assistance. The process is simple; just upload a file, select your options, and add a payment method. Wait a bit, and the best of them generate very usable transcripts without the headache. There are a couple of things to consider before choosing a transcription service, however.
Accuracy is the most important concern of all, and choosing the wrong type of service at the outset might leave you with a significant amount of editing to do. Cost is also an important factor. Although most transcriptions services charge on a per-minute basis, prices vary, and some services offer bulk plans at better values.
The cheapest (and probably most accurate) way to transcribe an audio or video file is to do it yourself. In other words, you listen to the audio file and type or dictate what you hear. You can do this with any number of programs, but it's often cumbersome to synchronize media playback to your typing speed.
Between laboring over every word, setting up the correct formatting, and reviewing the finished product, it's enough to steer most people towards a dedicated service. Automatic transcription services are the next step up from the manual approach. For this method, you upload files to a program that processes the audio quickly using automatic speech recognition (ASR) and spits back out a transcript.
The downside of automatic services is that they are far less accurate than other methods. Otter, Trint, Temi, and Scribie all offer automated transcription services—Scribie also offers a human-based service. With higher-tier transcription services, a trained transcriptionist (often more than one) completes the work on your file. These services are highly accurate, but they're also pricier and typically require a longer turnaround time.
Many transcription services charge on a per-minute basis. For example, a 30-minute transcription at $1 per minute would cost $30. Costs can quickly add up, and some services bill extra fees for a faster turnaround, for verbatim files (including all the "ums" and "ahs"), or if the audio is of poor quality ((⇨ view some video transcription services)).
Otter and Scribie's automated transcription tiers are the only free options we've reviewed, though the former will eventually require a monthly subscription. As you might guess, the amount of time it takes to turn around a file usually depends on its length. Automatic services can typically process a file in a matter of minutes.
Rev is simple in that it promises to return your file (in most cases) in a 12-hour timeframe. Scribie's and GoTranscript's slowest options (five days) are also their cheapest, though you can fast-track that to 12 hours for an additional $1.60 per minute in both cases. There are also intermediary options for both services.
It is vital that your subjects are close by and speak in loud, clear voices. If there are multiple speakers present for a recording, participants should only speak one at a time to avoid interference. Most services also point out that speakers with heavy accents may also pose some issues, though there's not much you can do to avoid that. (⇨ get youtube transcribed accurately here).
It's also worthwhile to use a dedicated digital voice recorder. In-person recordings also produce better results than recordings of phone calls. In our testing, the overall accuracy of transcripts varied considerably. We evaluated with two different files: a recording of a conference call with multiple speakers and an in-person interview with just two participants.
The latter did considerably better on the second (easier) recording, but they still weren't perfect. Keep in mind that your experience may vary, as we cannot control every variable in tests of human-based transcription services. Basically, the automatic services are only useful if your recording is on the simple side and you do not need the utmost accuracy.
Regardless of the service you choose, chances are that you will need to correct some parts of your transcript. As such, most services include a built-in editor for making these changes before you export the final document. Typically, these interfaces combine playback controls with a text editor. This is much more convenient setup, then say, switching between a document and audio player every couple of minutes.
Some include extra tools for highlighting selected parts of a transcript or editing the start time of the recording (video transcription). Playback speeds and quick rewind buttons (all controllable via keyboard shortcuts) are also fairly standard. GoTranscript is notably the only service that does not offer an online editor; your only option is to edit the exported transcript after it completes a job.
All offer both Android apps and iPhone apps. For the most part, these apps function as digital voice recorders, but they do let you order transcripts of the recordings directly from your mobile device. The drawback is that you can't import audio files or links the way that you can via their respective web interfaces.
Otter goes one step further than the others with excellent organizational features and the ability to edit transcripts on the go. If you want to avoid the transcription services entirely—for privacy reasons or to save on costs—there are alternatives. For doing your own manual transcriptions (you listen to the recording and type what you hear), Transcribe is a great option, at only $20 per year.
For those who don't want to spend any money, Google Docs may be the best solution. With Google Docs, you can use its voice typing feature to put words down on the page, which is certainly quicker than typing everything out. Another completely free option is oTranscribe, but it operates more similarly to Transcribe, with a similar layout and set of keyboard controls. (⇨ visit Way With Words).
Yes, transcribing can be a hassle and some services are costly, but the value of accurate and usable transcripts far outweighs these annoyances. At least one of the services in the chart should suit your needs; make sure to read our full reviews for help picking the right one. Do you use a service not mentioned here? Let us know in the comments and it may make the chart in our next update.
Photo: Rozette RagoThe human transcribers at GoTranscript returned nearly 100% accurate transcriptions in a couple of days and didn’t balk at recordings featuring heavy accents.If you need transcripts that come ready for publication, or a transcript of an audio file featuring speakers with accents, GoTranscript is the best choice. It’s one of the most readable and accurate transcription services we tested, as it consistently returned transcriptions that were nearly 100% accurate.
But the price is worth paying if you don’t want to spend time cleaning up transcripts yourself. GoTranscript (human) 97% 85% 97% 99% Scribie (human) 89% 90% 98% n/a Rev (human) 87% 90% 96% 78% Temi (AI) 73% 71% 73% 42% GoTranscript got high marks on a range of scripts and audio files, and in many cases produced the most easily readable transcripts from human transcriptionists.
The few errors included typing “part of” instead of “in part,” and writing “$1,440” instead of “$1,414.” On the pangram section, which featured phrases that contained all of the letters in the English language, GoTranscript was perfect. When we submitted the same script with intentional background noise, the transcription had only similarly minor errors.
But we found two spots where words had been replaced with “inaudible” or “unintelligible.” GoTranscript did get proper nouns like Mulholland Drive and Bala Cynwyd correct, but the service inserted “unintelligible” labels four times for other place names in the last section, which affected its accuracy score considerably (transcribe footage).GoTranscript is the only service we tried that was able to accurately transcribe a recording of someone with a non-American accent.
Scribie didn’t return a transcript to us at all, stating that the file was too difficult. GoTranscript (human) $0.90 Scribie (human) $0.80 Rev (human) $1.25 Temi (AI) $0.25 The additional accuracy of having a person transcribe your recording comes at a much higher price.GoTranscript is the second-least-expensive real-people service we tested: 90¢ per minute for the first 180 minutes of recordings you upload, with lifetime discounts if you upload more.