Synopsis
The InterPlex has previously relied on Riverside as a means of generating automatic transcripts for InterDialogues. As I am recommending that we leave Riverside (see 3.7 Riverside alternatives), it is necessary to identify another transcription program. Here I suggest a high-functionality temporary option with a paid subscription that will cover our transcription needs while I look into how to transition into a free but more technically complex solution.
1. Temporary alternative
Given its low cost, high functionality, and positive reputation for professional use, I propose WhisperAI as an alternative. Whereas Riverside is a slow means of producing faulty transcripts with limited editing options, WhisperAI is a dedicated transcription program with high operation speed, high accuracy, and a variety of editing options (e.g., options for adding speaker names, automatic timestamps, automatic translation, conversation summaries, etc.). I have personally gotten good use out of WhisperAI for my own projects and can confirm that it works well with both English and Spanish audio. My recommendation is therefore to replace our Riverside subscription with a team subscription to WhisperAI.
A standard subscription is 99/year. This plan comes with all the basic functionality we need, but it is limited to 120 minutes of content per month. Depending on the number and length of transcripts we need to generate on a monthly basis, that may not be sufficient. The next tier up includes unlimited transcription hours at 199/year. This may seem slightly expensive, but compared to other alternatives, most of which are at similar price points, I believe this will work well to cover our needs on a temporary basis.
2. Transition toward a free workflow
WhisperAI is powered by OpenAI’s open-source Whisper model, though it seems that the former is an independent brand not directly affiliated with OpenAI; WhisperAI has simply monetized and further developed the open-source code of OpenAI’s product. It appears that there are ways to take advantage of the Whisper model for free, but I haven’t learned these workflows yet, and according to my research they look to be slightly complicated.
Given my time constraints, I am not sure when I will be able to develop a free workflow as an alternative to paid services, but it is something I’ll be looking into further over the coming months. My suggestion is that we use WhisperAI temporarily until I have the time to develop a workflow using a free application of the Whisper model.
Related
2.2 InterPod production checklist
2.4 InterDialogue production checklist
3.1 Audiovisual recording best practices