About Recording and Transcription

The way a qualitative researcher transcribes interviews should flow from, or align with, the methodological paradigm chosen for the study, whether that paradigm is interpretivist, critical, positivist, or participative in nature (or a combination of these).

Before embarking on a transcription project, it's worthwhile to invest in the time and effort necessary to capture good audio, which will make the transcription process much easier. If you haven't already done so, check out the audio capture guidelines from Harvard Library's Virtual Media Lab, or contact a media staff member for customized recommendations. First and foremost, be mindful of common pitfalls by watching this short video that identifies the most common errors to avoid!

Helpful Texts

Software

  • Adobe Premiere Pro Speech-To-Text automatically generates transcripts and adds captions to your videos. Harvard affiliates can download Adobe Premiere in the Creative Cloud Suite.
     
  • GoTranscript provides cost-effective human-generated transcriptions.
     
  • pyTranscriber is an app for generating automatic transcription and/or subtitles for audio and video files. It uses the Google Cloud Speech-to-Text service, has a friendly graphical user interface, and is purported to work nicely with Chinese. 
     
  • Otter provides a new way to capture, store, search and share voice conversations, lectures, presentations, meetings, and interviews. The startup is based in Silicon Valley with a team of experienced Ph.Ds and engineers from Google, Facebook, Yahoo and Nuance (à la Dragon). Free accounts available. This is the software that Zoom uses to generate automated transcripts, so if you have access to a Zoom subscription, you have access to Otter transcriptions with it (applicable in several languages). As with any automated approach, be prepared to correct any errors after the fact, by hand.
     
  • Panopto is available to Harvard affiliates and generates ASR (automated speech recognition) captions. You may upload compatible audio files into it. As with any automatically generated transcription, you will need to make manual revisions. ASR captioning is available in several languages. Panopto maintains robust security practices, including strong authentication measures and end-to-end encryption, ensuring your content remains private and protected.
     
  • REV.Com allows you to record and transcribe any calls on the iPhone, both outgoing and incoming. It may be useful for recording phone interviews. Rev lets you choose whether you want an AI- or human-generated transcription, with a fast turnaround. Rev has Service Organization Controls Type II (SOC2) certification (a SOC2 cert looks at and verifies an organization’s processing integrity, privacy practices, and security safeguards). 
     
  • Scribie Audio/Video Transcription provides automated or manual transcriptions for a small fee. As with any transcription service, some revisions will be necessary after the fact, particularly for its automated transcripts.
     
  • Sonix automatically transcribes, translates, and helps to organize audio and video files in over 40 languages. It's fast and affordable, with good accuracy. The free trial includes 30 minutes of free transcription.
     
  • TranscriptionWing uses a human touch process to clean up machine-generated transcripts so that the content will far more accurately reflect your audio recording. 
     
  • Whisper is a tool from OpenAI that facilitates transcription of sensitive audiovisual recordings (e.g., of research interviews) on your own device. Installation and use depends on your operating system and which version you install. Important Note: The Whisper API, where audio is sent to OpenAI to be processed by them and then sent back (usually through a programming language like Python) is NOT appropriate for sensitive data. The model should be downloaded with tools such as those described in this FAQ, so that audio is kept to your local machine. For assistance, contact James Capobianco.

Equipment

  • Transcription pedals are in circulation and available to borrow from the Circulation desk at Lamont, or use at Lamont Library's Media Lab on level B. For hand-transcribing your interviews, they work in conjunction with software such as Express Scribe, which is loaded on Media Lab computers, or you may download for free on your own machine (Mac or PC versions; scroll down the downloads page for the latter). The pedals are plug-and-play USB, allow a wide range of playback speeds, and have 3 programmable buttons, which are typically set to rewind/play/fast-forward. Instructions are included in the bag that covers installation and set-up of the software, and basic use of the pedals.

Need Help?