News and Notes related to Digital Media Transcription, Analysis, and Captioning.
RSS icon Home icon
  • Creating a Transcript with InqScribe and Dragon NaturallySpeaking

    Posted on September 12th, 2014 Alex No comments

    When we’re asked about speech recognition software, it’s often from new users hoping to automate their transcription process. Although InqScribe doesn’t include any speech recognition technology, you can easily use InqScribe in conjunction with other speech-to-text software.

    Dragon NaturallySpeaking is the most popular speech recognition product on the market. “Dragon” has been around for over 17 years, with version 13.0 recently released for PC by Nuance Communications. For Mac users, there’s also an OS X edition called Dragon Dictate. In this article however, we’ll be looking at Dragon NaturallySpeaking 12 Home Edition.

    As you might guess, Dragon essentially converts your voice into text, also allowing for command triggers such as “Open My Documents” and “Delete Line.” Although true transcription automation isn’t quite there yet, you can use Dragon effectively with your own voice.

    Automatic Transcription

    Wouldn’t it be great if a program could listen to audio and create an exact text of the recording? Well, in reality it’s a little more complicated. Before digging into Dragon, it’s worth laying out some of the difficulties of automatic transcription:

    1. Computers have a hard time distinguishing a single voice from the noise and general messiness of most recordings. There are often ambient sounds or background chatter in even the most pure recordings.
    2. It’s difficult for speech recognition programs to remain accurate moving from speaker-to-speaker. Everyone’s voice sounds a little different, so programs like Dragon work best by training with and focusing on one voice.
    3. Even in ideal conditions, speech recognition isn’t perfect. It’s never going to be 100% accurate because it mishears you or maybe it just doesn’t understand a colloquial word. A lot of what we “hear” from language comes from context and nuances in our speech and body language. Speech recognition technology just isn’t advanced enough to pick up on such complex information.

    So, if you’re still interested in trying out automation, the Pro and Legal Editions of Dragon 12 support speech-to-text conversation from audio recordings (click here for a PDF comparison of these editions). This feature is intended to work only with your own, trained voice due to the limitations explained above. However, there’s nothing stopping you from trying out the voice recognition on multiple speakers.

    One possible transcription method is to run an initial pass with Dragon, and then use InqScribe to follow along and edit its mistakes. I’ll note that we haven’t tested this method, and I imagine the results will vary widely from recording to recording. Check out this video on how to transcribe from an audio recording with Dragon 12 Pro.

    Parroting a transcript

    The most reliable method recommended by InqScribe users is to “parrot” your audio source. This entails listening to your media file and repeating everything you’d like to transcribe out loud into Dragon. Parroting is not automatic, but to some it’s a welcome alternative to the keyboard gymnastics of traditional transcription. Here are the steps:

    1. Launch Dragon NaturallySpeaking
    2. Open up a new InqScribe document
    3. Load your audio or video file by navigating to Select Media Source>Select File
    4. Press play (feel free to adjust the play speed to suite your pace)
    5. Listen to and repeat out load everything you’d like Dragon to write into your InqScribe transcript
    6. Once your finished transcribing, make sure to review the text for errors

    I found myself using voice recognition to get down the general text of the transcript while controlling media playback and timecodes using InqScribe’s keyboard shortcuts and my mouse. This frees your hands to focus on media manipulation. It’s also nice to have the precision of keyboard shortcuts for inserting timecodes.

    Dragon Transcript Capture

    InqScribe and Dragon in action

    How well does parroting work?

    As a speech recognition software, Dragon does a pretty good job of understanding your voice. It will even improve the more you use it. There is a bit of a learning curve, which mostly comes from training yourself to be effective with voice commands and corrections.

    You’ll have to learn to regularly check Dragon for mistakes because, as hard as it tries, it’s never going to be 100% accurate. Honestly, many of the errors I encountered felt like my own fault- from stuttering or not clearly annunciating  my words. As any transcriptionist knows, language is a messy thing. This means you’ll have to watch Dragon’s work for errors, which can disrupt your workflow. Even after your dictation, you’ll want to budget some time to make sure homophones like “example” don’t wind up as “egg sample.”

    Factoring in these corrections, I found parroting with Dragon to be slower than manually typing the transcript. This will certainly depend on your typing speed and how proficient you are with Dragon’s voice interface.

    Be aware that prolonged sessions will leave your voice tired. This might seem obvious, but you’ll get physically fatigued much faster than by typing normally. For this reason, I don’t think I would recommend Dragon to knock out wordy transcripts. I know my voice tires easily, so if you’re used to talking all day, it might be less of an issue for you (hello teachers).

    Overall, I’m not sure that I would recommend Dragon to someone who has no issue manually typing a transcript. It just doesn’t offer a clear advantage. If, however, you’re plagued by slow typing, mental blocks, physical impairments, or just love the sound of your own voice, then Dragon can be a great tool. Just be willing to make it through a high learning curve before you can really start plowing through transcripts.

  • Guest blog: A Filmmaker Uses InqScribe for Translations and Subtitles

    Posted on May 22nd, 2014 christina No comments

    InqScribe is popular among filmmakers for a variety of reasons. In our latest guest blog, Scott Squire shares his workflow to translate and create subtitles for his documentary.

    by Scott Squire, Independent Producer & InqScribe user

    Our movie, The Girl Who Knew Too Much is a feature documentary about women’s empowerment, the complexities of globalization and the crisis of mental health support in Nepal. We shot our movie over seven years on six different cameras, with varying degrees of asset management rigor. We ended up with nearly 200 hours of footage to be logged and/or translated and subtitled. This project taught us the value of having solid workflows—from managing footage in the field through the process of translation. We cannot stress the importance of having a process (that works with every camera!) before you shoot a single frame.

    Our translators at work

    Our workflow includes the use of InqScribe and Final Cut Pro to produce translations and final subtitles for our movie. Critical to our process is having a clear folder/file structure that essentially depicts our workflow. I will refer to the folder/filenames used on our project, but obviously these would change based on your own project details.

    Folder structure to organize our workflow

    Set up

    We make sure that the top-level directory is accessible from all of our workstations. Our workflow begins with our editor placing callouts for the media that needs translating in the folder called In1. Fiona Highlighted Documents. Sometimes these are snippets from a clip, sometimes a whole reel. She reads our transcripts and highlights them, and then sends us scans of the highlighted pages as PDFs. (Unfortunately, on this project, we didn’t have InqScribe at the start, which means our original transcripts were in a variety of formats.)

    Review the translation and verify timecode in InqScribe

    We assemble all of the relevant translated files and then create an InqScribe document for each clip. Our workflow involves moving the InqScribe transcripts through a series of reviews by native speakers who verify the accuracy of the translation and the placement of timecodes. Timecode references should be at least at 10-second intervals (we recommend every 4-5 seconds). Be careful with InqScribe’s sensitive timecode selection so that you don’t timecodes out of order. And be sure to have opening AND closing timecodes in InqScribe or else your subtitles will run too long.

    Export XML file from InqScribe to Final Cut Pro

    Once the translation is complete and the timecodes are reviewed, we use InqScribe to export an XML file to be used in Final Cut Pro. It’s good to keep your XML template easy to access because you’ll be using it a lot. We keep ours at the top-level folder of our shared directory: /Master Wrapping Subtitle Template v1.3 (use this one).xml.

    Instructions for creating a custom FCP 7 XML export template can be found in InqScribe’s Knowledge Base.

    Very important is the naming of files. InqScribe will show the default name as “export.xml.” Click the “Choose” button to select a destination to save the file. Navigate to /Out2. Subtitle XMLs for Final Cut Pro > Interview XMLs from InqScribe. The name of the XML needs to match the file it came from. The best way to do this is to navigate to the original InqScribe file, select and copy the name (all but the “.inqscr” extension) and paste this into the pane of the export dialog. Double check your naming and destination—this is a big pain if you get it wrong. CONSISTENCY and ACCURACY in naming and filing are CRITICAL.

    If you’re lucky, when you click “Export,” the export window will just close and your shiny new XML file will turn up in the proper place. If InqScribe alerts you to a non-consecutive timecode you will need to go back and correct where needed. The error will only tell you approximately where to start fixing first. Cancel out and go back to check all of the timecodes where the error message is indicated.

    Import XML files to Final Cut Pro (FCP)

    In Final Cut Pro, you will create a project (Subtitle_Master_Project) and import the first InqScribe XML file in the scene you just created. When you import the InqScribe XML, it will come into FCP as a sequence with the generic name that matches your text generator template (Master Wrapping Subtitle Template…). You must change this back to the name of the original file. As before, navigate back to the file and copy and paste the name in the FCP browser to replace the template name.

    Drag the newly imported sequence into the timeline. It will contain only a subtitle track. Next, import the associated master footage or locate the clip and drag it into the FCP browser window. When the media clip has been imported, be sure it’s in the right folder inside the browser. Drag the clip into the timeline so that you have both footage (video and audio) AND a subtitle track above the video. These should be the same length, but there may be gaps in the transcription file. If the closing timecode was correctly input in the InqScribe file, the endpoint will match the end point of the media.

    If you’ve done your job in InqScribe, the clip lengths should all match up and you’ll have a subtitled sequence! You can now export the sequence from Final Cut Pro as a XML to send to your editor.


    About the author

    Scott Squire is a filmmaker who, along with his wife Amy Benson, is the producer of The Girl Who Knew Too Much.

  • Guest Blog: A Final Cut Pro Workflow for Editing in Another Language

    Posted on May 7th, 2012 ben No comments

    Finding the right workflow for a project can be a challenge, especially when you have multiple languages, multiple tools, and multiple collaborators in the mix.  In our latest guest blog, Chad Braham, an editor and Director of Media Production at WORDonCancer.org describes in very helpful detail how they’ve developed a translation workflow that starts in Final Cut Pro and ends back in Final Cut Pro with a full resolution subtitled version that enables him to edit the film in a language he doesn’t speak.

    Got an interesting story about how you’re using InqScribe? Please contact us at support@inqscribe.com if you’d like to highlight your work.


    by Chad Braham, Editor, Director of Media Production, WORDonCancer.org

    WORDonCancer.org is a 501(c)3 non-profit organization based in Indianapolis, IN, that aims to educate and raise awareness about women’s cancer.

    This summer, our organization is working on a documentary short film about a cervical cancer prevention program in the Peten region of Guatemala, entitled, “No More <http://nomorethemovie.com/>. Due to the location of where the film was shot, a large portion of the footage needs to be translated and subtitled (from Spanish to English) for the final piece.

    Much of this translation is currently being done by a small group of volunteers, most of whom have little to no subtitling/transcription experience. Because of this, we needed a solution that was easy to learn and use, was available on multiple platforms (PC and Mac) and could work within our Final Cut Pro 7 video editing workflow.

    As you probably already know, Inqscribe <http://inqscribe.com/> does all the above and more. As the editor of this film, I must say that I don’t speak much spanish, so it was critical that our translation workflow start and end with Final Cut Pro. With Inqscribe I can edit spanish speaking interviews, in Final Cut Pro, with subtitles, and find sound bites and edit points as if I were editing an english speaking interview. Our typical workflow is as follows:

    1. Export a clip of spanish speaking footage from Final Cut Pro as a small reference video file with a timecode window burned in. In an attempt to keep the physical file size small (and the duration short for our volunteers) we usually keep the file to around 7 minutes long (a 320×180 Quicktime file at Photo-JPEG with “Medium” compression seem to play-back better then “.mp4” files on slower PC machines).

    2. Upload the file to our FTP site or Google Drive (depending on the volunteer’s preferences) and notify the volunteer with an email that also includes a few notes about the file (who the person speaking is, why we chose to talk to them, etc.)

    3. Volunteer transcribes in Inqscribe and breaks up the transcription into phrases with timecode for subtitling exports.

    4. The volunteer then emails over just the “.inqscr” file, that the video editor opens in his copy of InqScribe, makes a few adjustments to ensure it adheres to basic subtitling best practices (amount of text per subtitle, etc.), and then exports out an XML from InqScribe for Final Cut Pro import.

    5. The XML export is then imported into Final Cut Pro and sync’d to the full-rez version of the translated clip. This is really the beauty of translating and transcribing with Inqscribe, the translation is already broken into full-resolution subtitling text “slides” in FCP and can be edited further without any quality loss to the video or text.

    One of the biggest challenges of producing any documentary, is organizing the massive amounts of footage and material, into a compelling story. This is especially complicated when a good portion of the material you are working with is in a foreign language. Luckly, there is a software like InqScribe that is so easy, anyone can use it.

    You can learn about this film at the film’s website www.nomorethemovie.com.

  • Charity: Water

    Posted on October 3rd, 2011 ben No comments

    We’re proud to be sponsors of charity: water.  InqScribe’s ability to support transcriptions and subtitles in multiple languages make it ideally suited for work in the field, especially for organizations such as charity: water that rely on story telling.  Here’s how they describe their work:

    Boy drinking water

    Photo by charity: water

    In 2006, we followed women in Ethiopia to a tepid water hole, where they filtered muddy liquid through their headscarves into a Jerry can. They hauled more than 40 pounds of this water on their backs to take home to their families. This is the only drinking water they had and they had to make it work for cleaning their homes, cooking, and unfortunately- drinking.

    We filmed this. We shared their stories to show what millions of people in developing countries do each and every day. The response was overwhelming.

    Fast forward five years and we’ve funded more than 4,200 water projects around the world to serve over two million people with clean water. Why? Because we’ve stuck to telling honest and beautiful stories from people living through the water crisis. As a result, thousands have joined our mission by donating or fundraising alongside us.

    Each story we tell empowers giving as we return and share it with those willing to help… but we currently do all of our post-production work with very little equipment. Since we’re shooting in places that require a great deal of translation and subtitles, we also need to do a lot of transcribing.

    InqScribe’s software allows us to quickly transcribe in one program — instead of having to toggle back and forth between a video and an excel spreadsheet. Not only that, the shortcut controls allow us to customize each document, thereby streamlining our editing process. Speeding up our post production only makes us more efficient in sharing stories that need to be heard.

    Thank you for your time and all you do to improve video storytelling.

  • Guest Blog: “This is definitely the fastest transcription we’ve ever done.”

    Posted on September 13th, 2011 ben No comments

    Filmmakers like InqScribe for a multitude of reasons. In this latest guest blog, part of our ongoing series highlighting how folks use InqScribe, Bongiorno Productions, an award-winning filmmaking duo, talk about how they’re using it with their latest project.

    Got an interesting story about how you’re using InqScribe? Please contact us at info@inquirium.net if you’d like to highlight your work.


    Screenshot from The Monks

    Monks in the Hood

    Emmy-nominated, award-winning, husband and wife filmmakers, Marylou and Jerome Bongiorno (http://www.bongiornoproductions.com), are in production on a new documentary called The Rule. It’s the story of Benedictine monks working in inner city Newark, NJ, as a successful model to combat the plight of urban America – to read more about the production, please see this news article.

    Filmmakers’ comments on InqScribe:

    Since we shoot a lot of footage when we’re creating a doc, including both vérité and lengthy interviews, there is no way to remember it all for editing. So, we log and transcribe the footage. We then read over the transcripts, highlight key sound bites, and edit the bites into a smooth story.

    This process requires InqScribe – an exceptional, flexible transcription software that allows you to:

    1. Customize controls on your keyboard to facilitate quick typing.

    2. Type directly into a user-friendly, neat looking console that plays back the video or audio file. Plus, the console is intuitive, meaning you can start transcribing almost immediately.

    3. Punch in timecode as you’re transcribing, without stopping the playback – a great feature. We used to transcribe by using an editing program (so that we can copy the timecode) and a word processing program. The time spent switching back and forth – copying and pasting timecodes – doubled the transcription time. Punching in timecode is now easy, and we do it frequently for a more efficient transcript.

    4. Start your timecode from zero, set a custom timecode, or the software can read the media timecode (essential for editing).

    5. Slow the audio or speed it up. InqScribe does both at custom rates.

    6. Export the user-friendly files to many different file types to facilitate reading or printing. Even if some footage doesn’t make the final cut, transcribing is essential in archiving everything into a searchable format for future use.

    Bottom Line:

    If you have some typing skills and your time is valuable, InqScribe is the “it” software. It makes transcribing fun and highly efficient. This is definitely the fastest transcription we’ve ever done.

  • Taking and Sharing Notes on Video

    Posted on May 11th, 2011 ben No comments

    InqScribe’s free-form text editing is intended to support a wide variety of tasks. So while we often talk of “transcribing” video, in practice oftentimes it’s more efficient to simply take notes. I was reminded of this in recent conversation with one of our customers at a cable tv network: sometimes all you want to do is review a clip and call out highlights from the video that you might want to use.

    This is particularly useful in team environments: you can watch a video, insert a timecode and quick note about interesting moments, and then share the InqScribe document with your editor or director. The editor/director can click on the timecode to view the segment of video.

    For example, we need to edit our introductory screencast. I review the video and I write the following in an InqScribe transcript to tell Matt (who’s editing the video):

    Cut [00:00:06.00] through [00:00:14.23] -- we can jump straight into the intro.
    The audio at [00:00:27.00] is unclear.  Can you re-record that?
    End the clip at [00:01:06.03].

    Matt already has the video on his computer, so I can just email the InqScribe file to him. He’ll copy it to the folder where the video is, open the InqScribe file, and the transcript will automatically link itself to the media again. (Alternatively, you can reference a common file on a network somewhere). He can then click on the timecodes I’ve inserted to see exactly what I”m talking about.

    The “transcript” is only few lines, but it conveys everything he needs to know. And obviously, there’s no need for a line by line transcript. A few timecoded notes suffice.