Since Transcriptive is a new product, we don't have many user examples. If you've used Transcriptive on a project, we'd love to see it and show it to the rest of the world. Contact firstname.lastname@example.org !
After helping to run several media departments in Orange County (CA), video maker Zach Dove decided it was time to break free and find new ways to assist other businesses in creatively communicating their particular message. He started ZD Productions in 2008 and has been continually working to further his knowledge of storytelling through the medium of motion/still pictures as well as graphic arts.
The company quickly evolved along the years, but its mission stayed the same. “We provide a place of collaboration where we each bring our distinctive talents to the building of a more powerful story that will motivate your audience to thought and action”, explains Dove. The main goal is to provide highly personalized videos in a short period of time, which requires a skilled team, a lot of dedication and constant investments in new technologies.
The ZD Productions crew shoots around 50 hours of footage every month and going through all the clips to find sound bites used to be very time-consuming. “We mostly shoot social impact videos for organizations and nonprofits so all of our videos are very heavy on the interviews. We use Transcriptive to transcribe the entire interview so that we can read through it, highlight the sound bites that we feel are going to be useful, and start assembling the story”, explains Dove.
The ability to get his footage transcribed almost instantly makes it possible for Dove’s team to start piecing together the story quicker, sometimes moments after they finish the actual interview. “It has seriously been an incredibly freeing experience to be able to quickly (and cheaply!) receive transcripts of our interviews. It speeds up the editing process too. As now, instead of trying to scrub through hours of interviews to find that one sound bite that we’re pretty sure somebody said on Day 2 of the shoot, we can simply search the transcripts for particular words or phrases and BOOM! There we have it”, says Dove.
For Dove and his team, Transcriptive has been a very efficient way to convert speech into text quickly and efficiently. Although he still goes back to the original footage to catch eventual Artificial Intelligence mistakes on the transcripts, Dove highlights that having transcripts time-coded inside of Premiere Pro in minutes is an incredible advantage. “Transcriptive allows us to spend more time on crafting the story and polishing the visuals and not waste hours on transcribing everything by hand or going back through the interviews looking for a specific sound bite. I Most definitely would recommend the plugin to anybody wanting to produce videos more efficiently”, Dove concludes.
Annalise Ophelian started making feature-length documentaries in 2008, fueled by a desire to see more stories of under-represented communities on screen. Her work focuses on themes of representation and self-determination, giving voice to female protagonists and putting women in central story roles. “ I became a filmmaker in large part because of the magic I felt watching Star Wars. (...) For many of us, Star Wars was the first time we saw a female character holding her own on screen, and the first time we had permission to geek out on a genre previously reserved for boys,” explains Ophelian.
This early memory of self-empowerment directly inspired Annalise to produce her most recent film Looking for Leia, “a documentary about women who found identity, connection, and purpose in their love of the galaxy far, far away.” Now in post-production, the movie has generated over 100 interviews in a 14 month period of time, and quickly transcribing all this material quickly became a hard task to accomplish. That’s when Ophelian decided to incorporate Transcriptive into her documentary filmmaking workflow.
Transcription has been a big part of Ophelian’s routine since the very beginning of her career. The filmmaker is also a psychologist with a background in qualitative research and has spent many hours glued to her headset and foot pedal, manually transcribing both research and film interviews. “On my last documentary, I manually transcribed about half of my material and sent the remainder out to various professional transcription services. This was incredibly time consuming and expensive,” Ophelian recalls.
The transcription routine she had been dealing with for years was about to repeat itself in Looking for Leia when Ophelian decided to research automated solutions and found Transcriptive. According to her, the biggest advantage of Transcriptive in comparison with other AI transcription services was the integration with Adobe Premiere Pro. The integration allows her to transcribe, edit the transcripts, search content and generate captions inside of Premiere.
Ophelian started post-production on Looking for Leia in August 2018, and Transcriptive allowed her to generate roughly 1,000 pages of transcript over the past six months. “Using Transcriptive for Looking for Leia involved a fraction of the time and cost of previous projects. The turn around time of the AI software is amazing and having timecoded transcripts for all of my interview material means I’m getting to review all my footage. It makes it infinitely easier to find things by participant or subject,” says Ophelian.
Documentary filmmaking generates a lot of material, and incorporating new software into an already established production workflow is not always easy. Ophelian recognizes there was a learning curve before she could fully take advantage of Transcriptive’s automated transcription, search, and text editing tools. However, she guarantees that learning how to best incorporate Transcriptive into the Looking For Leia post-production workflow was time well spent.
In the beginning, Ophelian’s biggest challenge was to choose an export format that works best for her and the Looking For Leia team. “Once I was able to nail down a format that worked best for me I created two master documents: A Word doc with an interactive table of contents that allows me to search all of my interviews for keywords or phrases; and a Scrivener project where I plug these transcripts directly into a script template,” explains Ophelian.
According to Ophelian, transcribing has been very straightforward and accurate depending on the sound quality and content. “Looking for Leia is about women in Star Wars fandom, so there are a lot of pop culture references and obscure characters that Speechmatics didn’t recognize. But I generally found the mistakes to be highly entertaining. In one interview, it transcribed Bail Organa (Princess Leia’s adoptive father) to Lady Gaga,” she says.
Transcriptive offers two AI services inside of its Premiere Pro panel: Transcriptive AI and Speechmatics. Transcriptive A.I. averages 97% accuracy, is priced at $0.12/minute, and allows you to pay for the transcriptions as you go. Speechmatics is about 95% accurate and costs $0.07 per min. On the transcribing side, Ophelian has a tip for new users: “I learned that periods of non-dialog would produce long sections with single words on each line. These were easy to grab and clean up, and I generally exported an mp3 of the audio to use as a reference when doing transcript clean up so I didn’t have to comb through my proxies.”
It didn’t take long for the Looking for Leia team to get used to Transcriptive. Before the plugin, the transcription side of her post-production would often take 2 - 3 months. With Transcriptive, they were able to generate 60 interview transcripts in 6 working days, which allowed Ophelian to focus on the creative side of her filmmaking work. “Because Transcriptive is so efficient and budget-friendly, I have more time and money to devote to other aspects of post-production. Like all things in post, find a Transcriptive workflow that works for you. Do one or two transcripts to start with and experiment with the export format, figure out what works best for you. And the team at Digital Anarchy are great with answering questions,” Ophelian concludes.
The Michigan MediaThe Michigan Media group has been producing audiovisual content for the University of Michigan (U-M) since 1993. Always committed to quality, they focus on producing uniquely creative projects at an affordable price, working closely with the different academic departments to understand their needs and help them to engage with their audiences. “We have 7 people working together to produce a wide range of video stories for the University of Michigan. From marketing videos to events’ coverage and short interview pieces, we often need to send content to the U-M Social Media and Marketing units to convey the U-M’s most important messages”, says the Associate Director of Michigan Media Chris McElroy.
The University of Michigan has more than 275 degree programs and receives approximately 15,000 students every year. Michigan Media is a unit within the Office of Communications, so staying in touch with the different departments and collaborating with other communication units is fundamental. “We often send content we filmed and edited, as well as marketing pieces we produced, to the Social Media team so they can share on the U-M profiles”, explains McElroy. This process requires the Michigan Media team to work quickly and send content before the final videos are ready. This is where Transcriptive comes handy.
McElroy and his team of editors, producers, and videographers started using Transcriptive in 2018. They were looking for an efficient way to find content and a faster way to send it to the other University of Michigan departments. Since then, producing marketing pieces in conjunction with the Social Media team became much easier. “We used to share content much less often. When we did share content, we'd send out for transcripts (which would take days or longer) or we would share video files - a laborious process. Transcriptive has enabled us to transcribe, find sound bites, and share content faster. After filming, we can quickly use AI to transcribe. Using the web app, it is easy to share the transcripts among a group who can then review for content decision-making on paid social media”, McElroy says.
Michigan Media has been using the Transcriptive.com standalone Web App to share files every week. The app allows users to send a transcript or video file from Premiere to Transcriptive.com for a producer, client, or assistant to make a paper cut using a web browser, like Chrome or Firefox. Once edited, the transcript can be sent back to the video editor in Premiere to continue working with Transcriptive and PowerSearch. “There are whole layers of work that would be facilitated by using Transcriptive, but we're only using a portion of its potential since we're doing mainly shorter marketing videos (anywhere from :30 - 4:00). This would be really helpful in longer form documentary work, editing interview programs, etc”, McElroy highlights.
Transcribing interviews also became easier with Transcriptive. When asked about transcription accuracy, McElroy says their experience has been very positive overall. “Acronyms are often wrong but that would happen even with a human transcriber. We in the video department at Univ. Michigan are very impressed and find Transcriptive very useful. Even producer/editors who in the past have opted not to work with transcripts are finding it very helpful. It helps to find sound bites much faster, share content faster, make subtitles faster - everything. After the modest initial investment - it pays for itself big time since we no longer have to send out for transcripts.”
Transcriptive is a license-based plugin. After buying the license, clients can install it on two computers, but can only use it on one computer at a time. Transcriptions are paid directly to the chosen Speech-To-Text service. Clients can choose between Speechmatics and a Transcriptive-AI as their Automatic Speech Recognition engine. Speechmatics charges $0.07 a minute, and the Transcriptive service is $0.12/min but usually more accurate. However, both services are very accurate, and for difficult transcriptions, you can use the Transcriptive-HumanTouch service with a trained, native language transcriptionist ($1.50/min).
The Paladin Media Group has a well-established history in media production. With an average total of 20 current clients and 28 years producing content all around the US, including abroad, from Charlottesville to San Francisco, from New Mexico to Minnesota, their video production and post-production teams work closely with clients to provide unique media pieces. Their goal is always to deliver high-quality media at an affordable cost, which can be a challenge in video production. Filmmaking requires time, attention to detail and expensive equipment and that is why The Paladin Media Team is constantly looking for high-end technology solutions to speed up their workflow and reduce costs while keeping up with the high standards they have set over the years.
“Our video editing process sometimes starts while filming”, explains Ashby Wratchford , Paladin Media Post Production Supervisor. “While our production team is exhausting camera cards, editors start transferring captured media, so we can start transcribing and editing the videos right away. We currently have 6 people working in our post-production team, and taking advantage of technology to transcribe the interviews, help us find content, and deliver precise transcripts to our clients is crucial. Indeed, Transcriptive is our go-to tool for making closed captioning.”
To complete all the steps required by video production in a short period of time, without sending costs to the roof, the Paladin Media Group post-production team relies on automated tools. After spending a lot of time and money generating human transcripts, last year they decided to try incorporating Transcriptive into their workflow, so they could reduce costs without sacrificing quality. The result was more time left to focus on deepening the relationship with clients and investing in creativity.
AI became part of Wratchford and his team’s workflow last year when they incorporated Transcriptive to their workflow to produce personalized videos for the UVA Health System. It is an endearing series of short documentaries that used video to close the gap between doctors and patients. Paladin interviews doctors as they talk about their career and present themselves to patients. Series also features patients’ stories that inspire. “All our current projects with the UVA Health System have been transcribed with Transcriptive. We were constantly shooting over 20-minute interviews that would take time and patience to transcribe. As an experienced finisher, I cannot imagine myself having to transcribe all this footage. AI has been an incredible time saver. Before Transcriptive, the transcribing process was often skipped.”
Wratchford has been making movies professionally since 1995 and was skeptical about the accuracy of transcripts initially. “Watson (IBM’s speech-to-text engine) was a very happy surprise. We interview people from different nationalities, with many different accents. So far the accuracy has been about 94%”. For him, it is still necessary to go back to the transcripts and make small corrections before sending it to clients or generating captions, but Automated Speech Recognition (ASR) is still incredibly worth it. “We have been transcribing with Watson for a year and have not yet needed to pay for the transcripts. I do want to try using Speechmatics, understanding it works more accurately. But so far Watson has met all our needs and expectations and its accuracy continued to improve”, says Wratchford.
Partnering with speech engines that offer accurate transcriptions has been a high priority to Digital Anarchy, the team behind Transcriptive. However, the precision of the transcripts will vary according to audio recording quality, language, people’s accent, etc. Adding Speechmatics to our Premiere Panel as another option for Speech-To-Text conversion increases the chances of precise transcripts because it allows clients to beta test and decide what transcription services will work better for their needs. Speechmatics offers 18 core languages, while Watson allows clients to transcribe in 8 languages.
According to the Paladin Media Group, automated transcripts have been making not only editors but clients happy. Because the transcripts come back within minutes it is possible for them to see the content in a very short period. “They are able to read and discuss the interview with us almost instantly, which makes them feel part of the process and us more confident to edit according to their comments and specific needs”. Also, the ability to search for content allows video editors to find exactly what their clients want to include in the video without having to spend hours watching the recordings to find content. Wratchford highlights metadata search as a valuable tool to give quick feedback to clients and speed up his video editing. “On any given day we may interview up to three subjects for a specific purpose, averaging a 23-minute interview. Weeks may go by when the client comes back and says, “did so and so mention this and can we make a video out of it?” My first reflex is to go to Transcriptive”.
In business since 1991, the Paladin Media Group has gone through many video format and industry changes. AI was not always part of their reality and old transcripts created by humans was a big part of the company’s video production process. “Automated transcription is new to our workflow. We often need to go back to old footage and use transcripts created by our staff. Importing transcripts they already have created and letting Transcriptive assign a timecode to each word is one of the biggest advantages of the plugin to us”, explains Wratchford.
The strategy adopted by Wratchford’s team proves that the advantages Transcriptive brings are not limited to Automatic Speech to Text services. After importing transcripts from the computer into Premiere Pro, Paladin’s post-production crew can quickly get an idea of what was said and plan how each segment will be cut. “We can click on the word, jump to the video, and play the footage right away. The matching between transcript and footage through the timecode is pretty accurate even when the transcripts are not generated through AI”.
In addition to using human transcripts to locate specific content, Transcriptive includes a text editing platform that allows video editors and producers to edit the AI transcripts. The idea is to allow human and machine learning to work together for quick, but high-quality results. As previously stated by Wratchford, “AI transcripts are not perfect. Paladin editors have been experiencing an average of 94% when sound is well recorded, and accents are not too heavy”. Having the ability to edit the text allows transcripts to fully match the interviews before clients can get access to it.
Since the incorporation of Transcriptive into their video production, the Paladin Media Group has saved time and money on the making of over 100 short documentaries for the UVA Health System. “Incorporating Transcriptive into our workflow has been incredibly important to speed up our video production turnaround time and lower costs.” Every year, Wratchford and his team need to browse through an average of 24 Tb of footage just for the UVA Health System, and Transcriptive has been keying on managing and editing videos all year around. “We are happy with the changes and ready to try other tools now offered by the Transcriptive Suite of products”.
When asked about his next priorities, the post-production lead of Paladin Media Group mentioned investing in collaboration tools as an important step to improve the workflow created with the acquisition of Transcriptive last year. “It is great to know new tools are being incorporated to Transcriptive. I think the web app is a very interesting idea and we’ll try the BETA version next. It may make sharing transcripts with clients a lot easier depending on how well it performs and how user-friendly the interface is”, explains Wratchford. He is also planning to ask all his video editors to try the app as a solution for transcribing, searching and collaborate with co-workers who are not using Premiere Pro. “Some of my colleagues edit on Avid or Final Cut, others use Premiere. I personally prefer Resolve so the App might be a game changer when its full version was released”, concludes Wratchford.
The Transcriptive App lets you get and edit transcripts! However, the real power comes from collaborating with your team or clients. Simply send a transcript or video file from Premiere to Transcriptive.com for a producer, client, or assistant to make a paper cut. Once edited, the transcript can be sent back to the video editor in Premiere to continue working with Transcriptive and PowerSearch, our most recent addition to the Transcriptive Suite of Products.
Kelley Slagle began directing, producing and editing independent films with Cavegirl Productions in 2004. Here she talks about her advantages with using Digital Anarchy’s transcribing plugin.
I first heard of Transcriptive while preparing a talk I was giving on documentary editing for Post|Production World at the NAB Show earlier this year (and ended up recommending [it] to attendees of the class).
We are using Transcriptive to transcribe all of our interviews for EYE OF THE BEHOLDER. We have 40+ hours of footage, and the idea of paying a premium for that much manual transcription was daunting. I am in the editing phase now. We are collaborating with a co-producer in New York, and need to share our ideas for edits and content with him, so he is reviewing transcripts generated by Transcriptive and sending us his feedback and vice versa.
The ability to get mostly accurate transcription (I would estimate 85%) versus completely accurate is fine for us, as we did not expect the engine to know proper names of characters and places in Dungeons & Dragons. We are able to find what we need by searching a text file or searching the metadata thanks to the feature of saving the markers in our timelines. As editor, I will now be able to find an exact quote that one of my co-producers refers to, or find something by subject matter, and [I] anticipate this will speed up the editing process greatly. And when it is time to do transcription/captions of our completed project, I will use the plug in again on final exports.
As I said, I have recommended and continue to recommend Transcriptive to producers and editors. I was thrilled to hear of it earlier this year, and so glad to have become a regular user.
Cavegirl Productions, headed by founders Kelley Slagle and Seth Polansky, has produced numerous successful projects since 2004, including 13 short films, a web series, and is the creative force behind 2014’s award-winning comedy feature, OF DICE AND MEN. Additionally, Cavegirl Productions is on target for releasing a feature documentary in 2018 - EYE OF THE BEHOLDER, THE ART OF DUNGEONS AND DRAGONS.
Kelley Slagle, Cavegirl Productions
Having acquired a significant deaf community at Willow Park Church, we needed to figure out a way to get our online videos available to them. At first I considered recording the American Sign Language (ASL) interpreter, but was told that some hearing impaired people do not understand sign language well. So instead, I have been using Transcriptive to subtitle our online videos for the deaf. I’m also working [with our lead pastor] on a series of podcasts and I’ve been using Transcriptive to produce transcripts of the podcasts in order to make them searchable. (Note: We’ve not release the podcasts yet.)
One of the most useful and time saving features of Transcriptive is the fact that our lead pastor was born in the United Kingdom (UK), but has done ministry in many parts of the world which has resulted in a very unique accent. While there are many speech-to-text engines available, none could understand his accent. Transcriptive works very well. This saves us a lot of time too, being that we used to spend many hours manually transcribing. The time we now spend on it is reduced significantly. Seeing the text highlighted as the track is played, and being able to make edits without your hands needing to leave the keyboard is brilliant!
Until I used Transcriptive, I couldn’t find a speech-to-text engine that worked well with two speakers in an interview situation. This has been particularly helpful and Transcriptive does an amazing job, and it reliably identifies the two speakers. Speaking of which, the transcripts have been very accurate. Which is significant since I haven’t been able to find an engine that understands [our pastor’s] accent until now.
I’ve been recommending Transcriptive to colleagues now, and will continue to do so in the future.
Chris Neufeld, Willow Park Church
Windy Films is focused on telling stories of social impact and innovation. We tell real stories with real people. Our backgrounds are in documentary film and have taken then sensibility into our commercial work.
We use Transcriptive for everything. The integration into premiere has been incredible. We’ve been getting transcripts done for a long time. The workflow was always a clunky; particularly to have [transcripts] in a word document off to one side. Having the ability to click on a word and just have [Transcriptive] take you there in the timeline is one of our favorites.
Recently, we completed a project for Planned Parenthood. The deadlines were incredibly tight and Transcriptive was quite helpful in keeping to our schedule. Trying to beat the senate vote on the healthcare repeal bill, we were editing while driving back to Boston, Massachusetts from Iowa. The fact that we could get transcripts back in a matter of hours instead of a matter of days allowed us to get it done on time.
The accuracy of the transcripts has been a mix. When the audio is good the transcripts have been very accurate. We tried Watson at first, but found it to be a bit too inaccurate. Swapping over to Speechmatics has been solid for us since.
Harvey, Windy Films
Oh My! Productions (OMP) is a small production house in New York specializing in documentary films. We’re using transcriptive to transcribe both sit down and on-the-fly interviews, recently it helped OMP with our new documentary film Ren & Luca. Now, depending on the quality of the audio, Transcriptive can even be used for the odd verite scene. Furthermore, we’ve used it on finished projects to create transcripts for closed captions and subtitles.
OMP can’t even begin to say how useful Transcriptive has been; or how much time it has saved us. The turnaround time to receive the transcripts is SO much faster than when we sent it to a service. Sequence markers are used to mark up our interviews, so when we’re searching for a specific word, or phrase, we can easily find and access them.
Transcriptive allows us to find replacements when a critical word in a sentence has interference. Our workflow has become much smoother once we’ve incorporated the Transcriptive markers into our projects. Now, we keep the markers window open and can hop to our desired areas rather than shuffling back-and-forth between our text document transcript and [Adobe] Premiere.
The transcripts are as accurate as our speakers (we have a subject who tends to mumble) and sometimes there is a little more tweaking that has to happen for those subjects, but Transcriptive has been a great tool for very clearly recorded material. We’ve had the best luck with Speechmatics [speech services]. The time it saves outmatches the amount of time you need for corrections.
Transcriptive has been particularly helpful with our new documentary Ren & Luca. We LOVE Transcriptive, and have been telling anyone who is willing to listen. It’s been the game changer for [Adobe] Premiere we’ve been waiting for.
Joy E. Reed, Oh MY! Productions
L’appart is a small post-production suite in Paris, France. We specialize in editing, grading and sound for high quality documentaries. Adding Transcriptive to our workflow has been a great time saver, particularly the search feature. It is amazing to be able to find a word in a clip instantly just by typing the word in the search bar. Also, we were pleasantly surprised, that with clean French, we were able to get 100% accurate transcripts.
Of course, we would advise Transcriptive to everyone, but we want to keep a competitive advantage. Therefore, we mainly just mention it to our clients.
Benoit Degornet, L'appart
Transcriptive has already become an essential part of my ingest workflow. It has allowed clients to review raw footage faster and at their own pace. Many corporate clients do not have time to sit down and review a long interview, but with Transcriptive they can easily review text in the early stages to begin building a script.
Transcriptive has saved countless hours of reviewing raw footage to find soundbites a client thinks they remember from a past shoot. I am able to quickly find and access these clips in interview footage without the need for detailed log notes ahead of time. Saving time and manpower at these early stages allows for budgets to be economically stretched. As a result, we can spend more time on color grading, graphics, and other essential finishing in post-production.
Anyone who works with interview footage or creates documentaries needs to incorporate Transcriptive into their workflow. It is like having the best assistant editor with you at all times throughout post-production.
Name Withheld, Transcriptive Editor