Free Transcribing Program That Doesnt Require Uploading
The research
- Why you should trust u.s.a.
- Who should get this
- How we picked
- How we tested
- Our selection: Temi
- Flaws but not dealbreakers
- Highly accurate transcripts from real people: GoTranscript
- The competition
Why you should trust us
I'g a professional journalist who has conducted more than than a grand interviews over the years. Although I usually transcribe my own interview audio files the complimentary and old-fashioned mode (typing quickly equally I play and rewind each sentence 100 times while cringing at the sound of my vox), I have tried numerous other options. Transcription services have always stood out as the most constructive. For this guide, in addition to in-depth testing, I read existing reviews of transcription services and consulted forums to find commonly recommended options.
Who should get this
Professionals and hobbyists who demand a text version of audio files—journalists, students, broadcasters, and across—can benefit from using a transcription service. Such services can cutting out hours of time spent manually typing up a transcript, and they get in easy to search through the contents of an interview, to find an sound sample in a large library of recordings, or to accept care of near of the work of transcribing quotes. Keep in mind that we tested the picks in this guide almost entirely with phone calls recorded with the TapeACall app. Information technology's possible that they perform differently with other types of audio, such as in-person recordings.
AI-based transcription services are a more than breezy, much faster, and significantly cheaper pick than services that utilise bodily humans for transcribing. Even the best AI services aren't perfect, but they are accurate enough to remind you of the gist of a recording and assistance y'all discover a specific part. That makes them useful for people who need a visual way to parse interviews, such equally journalists who record a lot of interviews, students who make coincidental recordings of their classes, or professionals who need to retrieve the contents of a meeting. Journalists also need to double-check quotes no thing what, so information technology tin make sense to pay less and go with an AI-based service. But if you take the AI route, y'all'll need to spend fourth dimension cleaning upwards the text. Skip the AI-based services in favor of a service employing real people if you plan to publish an entire transcript or need a completely accurate text file for utilize in a professional setting.
The best human-powered transcription services are most 100% authentic and struggle only with highly specialized language such as street names, which makes them more than appropriate for someone who wants an exact tape of what is said in an audio file. Podcasters who want a full transcription of an episode, professionals who need a thorough record of a meeting to distribute across a company, or journalists preparing a long Q&A article might find that they can salve a tremendous corporeality of fourth dimension using a human transcriptionist. Only be prepared to pay a lot more for the actress accurateness compared with AI services, and expect a few days of turnaround time unless you're willing to pay even more to get results quicker.
How nosotros picked
We scoured forums frequented by journalists, writers, and podcasters to find a range of commonly used transcription services. We also read reviews past Poynter, TechRadar, and PCMag. To decide which services to test, we considered the following criteria:
- Readability: The single most important factor in choosing a transcription service is the readability of the resulting transcript, so nosotros checked samples for grammar and spelling. We also researched services' self-reported accurateness and other people'southward experience with them. Amid human-backed transcription services, we favored those that claimed to exist at least 99% authentic. Many AI-based services don't report accuracy, but ane option we tested claimed to exist at least 90% accurate for clear sound.
- Cost: Whether your employer or customer pays for your transcriptions or information technology comes out of your pocket, it'due south important that they be price-effective. Nosotros found that the near commonly recommended AI services price effectually 25¢ per minute of sound, and services employing man transcriptionists cost up to $2 for a minute of clear audio.
- Turnaround time: AI-based services take a maximum of i to 2 minutes per infinitesimal of submitted sound to return a transcription, merely human transcriptionists can take days to deliver a transcript. If a long delay volition interrupt your workflow or cause you to miss an important deadline, an AI service is far more affordable than a rushed transcription from a person. Because nosotros saw such a broad range of turnaround times, we considered the advertised deadlines for AI services and existent-people transcriptions separately. For the latter, we considered merely those transcription services that could return a transcript within a calendar week, and we took special notice if they promised a turnaround time of hours or even minutes.
- Support for complicated audio: Not all audio files are recorded in a professional studio, so the all-time transcription services should be willing or able to handle background noise, jargon, and accents. Although we tested only with recordings of English speakers with American and French accents, we likewise noted what languages and accents each service claimed to support. Some services, such as Temi, state explicitly that they do not support languages or accents other than American English, while others, like Trint, offer transcriptions for multiple languages and accents. We prioritized American English language for this guide.
- Transcription editors: The best services keep an online database of your transcriptions indefinitely. They also give you an in-browser space or mobile app where you lot can piece of work easily in a make clean pattern, editing text, listening to audio playback when y'all click anywhere within the text, and changing the speed of the audio. AI-based services are so error-prone that it'southward overnice to have the ability to jump in and edit a transcript on the service'southward website. Because human transcriptionists are so accurate, we didn't consider it as important for those services to include as many online editing features.
- User interface: Websites that allow you to upload audio files intuitively and chop-chop make the overall transcription procedure more than pleasant. We preferred services that laid out extras, such equally adding time stamps or selecting a faster turnaround fourth dimension, on one folio and besides showed the price of those add-ons. We dismissed services that obscured the toll until after we placed an order, made us pass through many pages to identify an club, or had a messy appearance.
- Security and privacy: Although nosotros don't recommend uploading sensitive audio files to a transcription service, nosotros even so looked into the policies each visitor had for protecting customers' data. Nosotros examined what type of encryption the companies used and whether they offered NDAs, and we looked for other security measures they took.
How we tested
Because nosotros institute that readability was the virtually important factor, nosotros tested each service with a variety of conversations and levels of sound quality. Nosotros wrote two scripts based on real interviews that reflected the different types of content a good transcription service should be able to handle:
- Our 237-discussion "control script" discussed drones, included common vocabulary and numbers, and concluded with a series of pangrams—phrases that apply all the messages and sounds of the English language language. Information technology also featured an pause where one speaker talked over the other.
- Our 172-discussion "jargon script" included jargon about batteries, particle physics, and place names.
Ii professional actors recorded themselves reading both scripts over TapeACall, an iPhone recording app that journalists normally employ. I also recorded a woman with a French accent reading the entirety of the control script over TapeACall. Overall, we made four recordings:
- Control script (clear): The actors read the control script clearly and without groundwork noise to test the all-time possible outcome from each of the transcription services.
- Control script (with groundwork noise): The two actors read the command script while music played and dogs barked in the groundwork to test the services' ability to pick up on the audio that mattered.
- Control script (with emphasis): The woman with a French accent read the command script to measure which services did better with not-American accents.
- Jargon script (articulate): The actors read the jargon script clearly and without background noise to test how the services handled unusual words.
During a second round of testing in 2020, we added another audio sample: We recorded a coming together with the iPhone's Voice Memos app, a sample intended to test whether the services' performance would modify dramatically when an sound file was fabricated with an app other than TapeACall.
We submitted all of the audio samples to each transcription service and recorded how long it took to upload the samples, plus our experience navigating the user interface. We timed how long each service took to return the completed transcript and compared the transcript confronting the original script to gauge accurateness.
We measured the quality of the transcripts in 2 ways. First, nosotros read each transcript and ranked how easy it was to understand compared with the other services' transcripts, a factor that we labeled "readability." And then we counted the number of words that were correct and divided that amount by the full number of words to generate a percent that reflected accuracy. Although it'due south easy to depict a conclusion from that pct, at that place'southward a meaning divergence betwixt a service that transcribes "batteries" every bit "battery" and one that transcribes "batteries" equally "basketball." So although raw accuracy is important, nosotros ranked readability as even more important.
We likewise recorded the total cost for each transcription and tested each service's editor to determine how easy (or difficult) it was to use. Finally, we checked to see whether each service offered helpful options such as time stamps, the ability to add names, and a place to submit vocabulary words.
Our option: Temi
Our pick
Temi
The best transcription service
Within minutes, Temi returned transcripts that were easier to read than what other AI services produced, even when the sound file wasn't perfect or when the words were difficult to follow.
Temi is the best choice for fast, inexpensive transcripts. In our tests, it beat out other AI-based services in readability and accurateness, and it returns transcriptions within minutes or hours instead of the days that a human transcription service typically requires. When the accuracy isn't perfect, its web-based editor lets you hands jump into a transcription and adapt the text or listen to the synced audio. Information technology'due south besides the second-to the lowest degree-expensive AI selection nosotros tested, then it'due south an affordable choice for freelancers or anyone paying out of pocket.
AI-based transcription results of English pangrams in 2018
| Original pangram | "The biscuit hue on the waters of the loch impressed all, including the French queen, before she heard that symphony again, simply as immature Arthur wanted." |
| Temi transcription | "the face up you lot on the waters of the lock and printing all including the French queen before she heard that symphony once again, just as young Arthur wanting" |
| Otter transcription | "the face up you on the waters of the lock and press all inclusive French queens before she heard that Symphony confronting simply a immature Arthur desire" |
| Trint transcription | "the space shoe on the waters of the lock and perchance all a French queen before she heard the symphony once more. Just did Arthur desire" |
The AI-based transcription services struggled with pangrams, which include unusual phrases. Although Temi's transcription was far from perfect, it was the easiest to read. The transcriptions featured above are from the clear recording of the control script.
In our 2022 testing, Temi returned a more readable transcription than Otter and Trint—the two other AI-based services we tested—for all four test recordings, and it as well had the highest accurateness pct beyond the board. The first part of Temi'southward control transcript, which discussed drone regulations, had accurate grammer and was highly readable. The 2nd part, which was made up of pangrams, had more than inaccuracies. In total, it was 73% authentic.
AI-based transcription results of jargon-filled speech in 2018
| Original script | "Researchers make neutrino beams by accelerating positively charged protons and smashing them into beryllium or carbon. This produces pions and kaons." |
| Temi transcription | "researchers make new train of these past accelerating positively charged protons and nifty them into glucinium or carbon that's produced is pions and cans" |
| Otter transcription | "researchers make neutrino these by accelerating positively charged protons and groovy into beryllium or carbon this produces pions encounter" |
| Trint transcription | "researchers make neutrino beams by accelerating positively charged protons and slap-up them into a volume of carbon. This produces Tientsin can" |
We constitute Temi'due south jargon transcript to be the nearly readable, only that doesn't mean it was all that authentic. In this case it mistook "neutrino beams" for "new train of these."
Temi still performed meliorate than the competition when we added background noise to our test recordings. In this example Temi'south AI fabricated only a few more than errors than it did on the control, which means information technology's still worthwhile to use Temi fifty-fifty when your sound isn't completely clear. Temi's jargon transcript was even more accurate than its work on the control; the commencement part of the jargon transcript discusses batteries, and we idea Temi'southward result on that was the most readable of the AI services' transcripts. The final section includes difficult place names such as La Cienega, Wayzata Boulevard, and the Schuylkill. That part stumped Temi—every bit it did even the human transcriptionists. Finally, Temi's transcription of the meeting recording we fabricated in 2022 had simply a single error.
Toll comparison (AI services vs. human-transcription pick)
| Service | Price per minute of audio (clear audio with no add-ons or discounts) |
| Temi | $0.25 |
| Otter | Free (for first 600 minutes each calendar month with limits) |
| Trint | Subscriptions get-go at $48/month |
| GoTranscript (our human-transcription pick) | $0.90 |
For longer files, at that place's a large cost divergence between our favorite AI service, Temi, and our favorite existent-people service, GoTranscript.
Temi charges 25¢ per minute of uploaded audio, tying for the cheapest option we tested (aside from Otter and YouTube, which are free simply wildly inaccurate). Trint's unlimited subscription-based model is cheaper merely if you upload at least 240 minutes of audio a month. The algorithm behind Temi doesn't care how complicated your audio is, so the price is ever the aforementioned regardless of what you submit. Human transcription services, meanwhile, start at fourscore¢ a minute and go up based on sound complexity and added features. Temi took between two and five minutes to render each transcript, which ways it took 1 to two minutes to transcribe each infinitesimal of sound. There are expensive human options that can return a file inside hours, only most not-AI services—including our pick, GoTranscript—take at to the lowest degree a few days to evangelize results.
Temi had the second-best editor nosotros tried in terms of our ability to bank check for and correct errors. Trint and Rev share the same editor, which we ranked as our favorite because information technology looked nicer, though it didn't take any actress features. Temi's website stores all of your transcripts indefinitely and allows you lot to edit them within a web browser, letting y'all piece of work with the text and the sound at the same time. We adopt that to the organization for some other services, which requires you to edit text in an function-software certificate and separately control an audio actor. With Temi, you can click anywhere in the transcript to hear the sound for that segment and blazon in your corrections. You tin slow down the audio playback or striking a button to rewind five seconds. The interface also offers tools for highlighting and striking text. When you're done, y'all can download the transcript as a PDF, Word document, or text file, or share it via email or a link.
Temi had the 2nd-fastest average upload time of any of the services we tested. You take just 2 pages to navigate: an upload page and a payment page, neither of which requires you to submit whatsoever additional data about the file. The user interface is modern and clean, and then uploading files and editing transcripts is intuitive.
Temi says it stores and transmits information using TLS 1.2 encryption, which we recollect is plenty secure, and you tin can asking a non-disclosure agreement. Temi says no humans view your information unless you share a link to your transcriptions. You tin also cull to delete text and audio files from the site in one case you've downloaded them; nosotros recommend doing then if you consider a recording sensitive.
Flaws just non dealbreakers
Temi doesn't allow you to submit speaker names or jargon in advance (though y'all can edit them after the fact with the company's editing tool), a characteristic that other services offer to ameliorate their transcriptions. It likewise does not save payment data; in our tests, it required us to resubmit those details each fourth dimension we uploaded an audio file. One workaround is to use PayPal or preload your account with funds, which will allow you lot to check out faster. Yet, even when we had to reenter credit card information, going through the upload and payment process took u.s.a. less than a minute—similar to our feel with the other services we tested. You can also upload multiple files at once to cutting down on the number of times y'all need to bank check out.
Readability rankings (AI-backed services vs. our human being selection)
| Control | Jargon | Background | Accent | |
| Temi (AI) | 4 | 4 | 4 | 3 |
| Otter (AI) | half dozen | half dozen | half dozen | 5 |
| Trint (AI) | 5 | 5 | v | 4 |
| GoTranscript (human) | one | ii | 2 | one |
Temi produced transcripts that were consistently the easiest to read among the AI options we tested. In this table, "1" indicates the best service for readability, and "half dozen" represents the worst.
Temi's other flaws are mutual to every AI-based service—the merely way to avoid these problems is to pay substantially more than for a person to exercise your transcription. First of all, even though Temi beat other AI-based services in readability and accurateness, information technology yet returned transcriptions filled with errors. You lot should treat Temi as a way to help you detect a specific identify in an audio recording or retrieve the overall content, not every bit a service that perfectly prepares quotes for publication.
Like the other AI services nosotros tried, Temi specially struggles with audio containing accents. Since it doesn't promise support for languages or accents other than American English, nosotros weren't surprised to run into information technology struggle with our speaker's French accent. If you demand a most perfect transcription of a file with accented speakers, we recommend paying more for transcription services done past a existent person.
Temi's speaker detection and fourth dimension stamping are poor, which is a trend we saw beyond all the AI services. It returned our transcripts every bit one long paragraph attributed to "Speaker 1," even though iii of our submitted recordings featured 2 speakers. Its transcripts accept an initial "00:00" fourth dimension postage but no further annotation. Clicking anywhere in the text gives you a time postage for that department; for a visual representation, all the same, be prepared to do your ain formatting.
Highly accurate transcripts from existent people: GoTranscript
Also great
If you need transcripts that come set up for publication, or a transcript of an audio file featuring speakers with accents, GoTranscript is the best pick. It's one of the nigh readable and accurate transcription services nosotros tested, as information technology consistently returned transcriptions that were nearly 100% accurate. Services that employ human transcriptionists take days to render transcripts, in dissimilarity to minutes for AI-based services like Temi, and they are significantly more expensive. Simply the price is worth paying if you don't want to spend time cleaning up transcripts yourself.
Accuracy tests for unlike scripts and audio-quality levels in 2018
| Command | Jargon | Groundwork noise | Emphasis | |
| GoTranscript (homo) | 97% | 85% | 97% | 99% |
| Scribie (human) | 89% | 90% | 98% | n/a |
| Rev (homo) | 87% | 90% | 96% | 78% |
| Temi (AI) | 73% | 71% | 73% | 42% |
GoTranscript got loftier marks on a range of scripts and audio files, and in many cases produced the most easily readable transcripts from human transcriptionists. Scribie refused to transcribe our French-emphasis audio file.
When transcribing our control script, GoTranscript produced the fewest errors of any service nosotros tried. The few errors included typing "part of" instead of "in part," and writing "$1,440" instead of "$i,414." On the pangram section, which featured phrases that contained all of the letters in the English language language, GoTranscript was perfect. When we submitted the same script with intentional background noise, the transcription had only similarly small-scale errors. Two words were replaced with "unintelligible," a mutual tactic we saw from human transcriptionists to avoid inserting wrong words; this approach makes the problem areas especially like shooting fish in a barrel to spot and so that y'all tin leap in and edit the transcript yourself.
The first 2 parts of the transcript from the jargon-filled recording, which was slightly less accurate overall than the human-performed work from competing services, had merely a few inaccurate words. But we found two spots where words had been replaced with "inaudible" or "unintelligible." GoTranscript did go proper nouns like Mulholland Drive and Bala Cynwyd right, but the service inserted "unintelligible" labels four times for other identify names in the last department, which affected its accuracy score considerably.
GoTranscript is the only service we tried that was able to accurately transcribe a recording of someone with a not-American emphasis. At 99% accuracy, the GoTranscript transcription of our French-accented audio sample was the most accurate transcription we received overall and easily beat Rev's 78%-authentic transcript. Scribie didn't return a transcript to us at all, stating that the file was too hard.
Price comparing (homo transcription vs. our AI pick)
| Service | Price per minute of audio (clear sound; no upgrades or discounts) |
| GoTranscript (man) | $0.90 |
| Scribie (human) | $0.80 |
| Rev (human) | $i.25 |
| Temi (AI) | $0.25 |
The boosted accurateness of having a person transcribe your recording comes at a much higher price.
GoTranscript is the 2d-least-expensive real-people service we tested: ninety¢ per minute for the start 180 minutes of recordings you upload, with lifetime discounts if y'all upload more. It charges extra for files featuring groundwork noise or accents, which meant we paid almost 4 times the cost of Temi, overall, to use GoTranscript to transcribe our 5 test recordings. However, there's no fashion effectually paying more if you want the accurateness of man transcription. Multiple services offering trial credits or coupon codes, and GoTranscript gives you $10 of costless credit to starting time.
Transcription turnaround time in 2018
| Control | Jargon | Groundwork noise | Accent | |
| GoTranscript (human) | 1 day 22 hours | 1 day 22 hours | 1 solar day 22 hours | 1 twenty-four hours 17 hours |
| Scribie (human) | three days 8 hours | 2 days ix hours | 3 days 8 hours | n/a |
| Rev (homo) | 8 minutes | two hours | 35 minutes | two hours |
| Temi (AI) | 4 minutes | ii minutes | 2 minutes | 5 minutes |
| Otter (AI) | Nether a infinitesimal | Under a minute | Under a infinitesimal | Nether a minute |
| Trint (AI) | Under a minute | Under a minute | Under a minute | Under a minute |
Accurate transcriptions, done by real people, have fourth dimension. Scribie refused to transcribe our French-accent audio file.
If you're on borderline and you need highly accurate transcripts speedily, yous need to either pay GoTranscript the premium for rush processing or go with i of its competitors. To become the cheapest price, we selected the slowest possible turnaround fourth dimension: v days. You lot tin can choose turnaround times equally fast as six to 12 hours for a fee. GoTranscript took between i twenty-four hours 17 hours and 1 day 22 hours to render our transcriptions, just longer sound files could require the full five days. Scribie took ii to three days to return our transcriptions, simply Rev easily beat GoTranscript on turnaround time by giving us our files within hours. All of the AI-based services were even faster. Merely we retrieve it's worthwhile to wait the several days and get a more accurate transcript if you have the time.
GoTranscript's editor isn't the all-time of the services we tested, only because its transcripts take then few errors, you lot can expect to spend less time using information technology than you would with other services. Although it lacks features that competitor Rev includes, such as highlighting and read-along options (similar to how a karaoke motorcar highlights the words as you go), it makes up for that with its simplicity and ease of utilize. You can click anywhere in the text to play back that part of the audio and make changes. In our tests, the transcriber accurately identified unlike speakers, and each fourth dimension the speaker changed, a new paragraph began and the text was conspicuously marked with a time postage stamp (an option we paid for). The other man-transcription services also did this accurately, while none of the AI-based services were able to.
The upload process is simple: After you send an audio file, GoTranscript asks you to select details about the recording, including the number of speakers and whether the audio is low-quality or features accents. Y'all can as well select options such as time stamping or captions. It's clear when extra charges are involved, and the form includes a spot to submit speaker names or special terms so that you can help the transcriber improve their piece of work.
GoTranscript makes a few promises related to security. The company says information technology uses ii,048-fleck SSL encryption to transfer and store data, which nosotros consider secure enough. GoTranscript requires transcriptionists to sign a non-disclosure agreement, simply you can as well submit your ain agreement for them to sign. Audio files are also chopped into pieces 5 to 10 minutes long and spread among different transcriptionists and so that no one person hears the entirety of a recording. After the transcription is complete, GoTranscript deletes the recording from its system, though you can still access the transcription on its server. (You tin can delete it later downloading, which we recommend for sensitive files.) GoTranscript too offers specialists for transcription projects in sensitive industries, such equally medical and legal.
If you're thinking most using a human-based service like GoTranscript, it's worth considering the low pay that transcriptionists generally receive. GoTranscript'southward competitor Rev has been in the news recently for its depression wages, but GoTranscript's Glassdoor page is also total of complaints nigh low pay. You should also consider whether the recordings you are submitting could be disturbing, and whether y'all'd be subjecting a person to an unexpectedly traumatic experience at work.
The competition
AI-based services
Trint is a well-known AI-based transcription service, and information technology had the all-time editor and the fastest turnaround time of any service we tested. It besides advertises its ability to transcribe in multiple English accents and 12 European languages. However, in our tests it was less accurate than Temi, producing poorer readability on all iv sound samples while also being more expensive. Trint operates nether a subscription model that starts at $48 per month for ane user and 84 files per yr, with an unlimited option at $lx per month. If you lot're uploading at least 240 minutes of audio a calendar month or working on a team, this unlimited tier might be a ameliorate option for you based on cost.
Otter offers 600 minutes of free transcription per month (equally long as you record straight from the app or Zoom), which makes information technology an attractive option for anyone who wants to transcribe and sort lots of interviews on a nonexistent upkeep. Although our 2022 testing showed Temi to exist much more accurate, Otter trounce Trint in some cases. However, when information technology came to readability Otter consistently scored final, with transcriptions that sometimes read equally gibberish. It's easy to use, and information technology has the fastest upload times and a decent editor.
If yous're looking for a gratis transcription option, you can likewise try YouTube: Turn your audio recording into a video, upload it to YouTube, and then use the website's captioning service to generate a transcript at no cost (exist sure to set the upload to individual for security reasons). The YouTube upload process took and so much piece of work and time, still, that we chop-chop disqualified this option. If you want a free transcript, y'all're meliorate off using Otter.
Man transcription
If you need the accuracy of a real person doing your transcribing just have only hours of turnaround time to spare, Rev could be a expert selection. Rev has the all-time editor tool (in fact, it's the same editor that the AI-based Trint uses) and the easiest upload process of any of the homo services nosotros tested. But although it was more accurate than whatever of the AI-based services we tried, information technology consistently returned the hardest-to-read and most error-filled transcriptions (aside from the jargon transcription, on which it tied for the most authentic) while also being the costliest of the services we tested. The Rev transcripts were still readable, but we think it's worthwhile to wait a scrap longer for transcripts from the cheaper and more accurate GoTranscript service, if you lot have the time to spare.
Scribie took the longest of whatever tested service to return our transcripts and had the worst editor, the slowest upload procedure, and the poorest user interface. When we submitted our sound sample of a speaker with a foreign accent, Scribie rejected it; a client service representative stated that the file was too short and too complicated for the service to find someone willing to transcribe it. Scribie rejected a longer accented file, too. If you demand to submit an sound file only on occasion or have lots of clear audio files, Scribie could nonetheless be a good option—it's the to the lowest degree expensive real-people service we tried, and it produced easy-to-read and accurate transcripts. Only steer clear if you want to be sure that your uploads will exist accepted every time.
Source: https://www.nytimes.com/wirecutter/reviews/best-transcription-services/
0 Response to "Free Transcribing Program That Doesnt Require Uploading"
Post a Comment