interviewScribe 4+

I was initially put off after reading several poor reviews, then heartened by one of the longer, more detailed reviews, so I tried it and it did the job.

This app is designed to help humans do manual voice-to-text transcription (the human listens to a recording, and types in what he/she hears). It does not do automatic speech recognition such as the Dragon products claim to do. For it’s intended purpose it has pretty much all the functionality it needs, but it’s hampered by a confusingly-written manual and a number of poor user interface issues.

Basically: you create a new document, you “assign” an audio recording (behind the scenes it stores a low-fidelity copy of the audio in the document), then it automatically tries to break the recording into s series of separate spoken segments, or “phrases”, then you play each phrase, one at a time, and type in what you hear. You can re-play the phrase as often as desired, you can adjust the start and end points of each phrase, you can join adjacent phrases together or split a phrase into two parts, and you can create a list of names from which you can associate a name (usually the person talking) with each phrase. Later you can export all the phrases in a text document, with the times and speaker’s-name listed for each segment. So the functionality is good, and is pretty much everything you could want for this task.

But the user interface puts lots of obstacles in your way: You have to keep entering and exiting edit mode for each phrase to switch between typing text and selecting phrases. You can’t re-play a certain point in a phrase while you’re editing it: you have to switch out of edit mode before clicking at a random point in the sound timeline. To export everything as a text file you have to: go to the Dictated Text tab (which is largely empty), click a plus button to create a new “variation” in a list of variations, then select a variation from the list, then click an Open button, then click a “Join Phrases Text…” button, then select some settings from a menu that pops up, then clicking a “Join” button in the popup, then review the combined text, then click an “Export…” button to finally create a text file with all your text.

Sounds like a nightmare — but it’s worth it. Manual voice transcription is a long and arduous task, and using this program, even with all its warts, will save you a ton of time.

The only other manual voice transcription app I know about for the Mac is “Transcriva” from Bartas Technologies (not available in the Mac app store). I have found this program (Interview Scribe) to be far more useable.

I am in an immersion-based Spanish language learning program in Mexico. I have hours and hours of raw language data in the form of digital texts and stories of varying lengths. Part of my studies incudes transcribing many these texts and stories for analysis. I have to transcibe the data myself, so that I am clearly understanding every single word in the text. I then check my transcriptions with a live language helper. Before I would just use iTunes to transcribe the texts, but I was always getting behind and having to rewind and try to find where I left off because I couldn’t type fast enough. This meant a lot of wasted time. So this program has been awesome for my goals. It has saved me a ton of time. The only thing I would change is to add a “loop” feature that allows the segment to keep repeating itself for the times I cannot type fast enough or when I need to hear it over and over to discern what is being said. Anyways, its the best app I have encourntered for my needs.

This app is a bit rough around the edges, but it does what I need at a great price.

I am not into any professional level transcribing, but I have hours of interviews of my relatives I would like to transcribe. It's been fun for me to go back and listen to a recording of someone I love who has passed away and type out their stories for my kids and others to enjoy. Sure, there are things that can be improved, but this app does that pretty well.