I write for a living and I record all my interviews. Temi has saved me weeks of transcription time -- and at prices I can afford. Temi is very good at transcribing different accents. I get 80%-90% accuracy. The read-along text cued to the recording make cleanup a breeze. I'd like to see some sort of payment tracking in my dashboard, so I can keep better record of my production costs. In a single word, indispensable.

Best automated transcript service I have ever used. Some of the more esoteric references stump the software, but overall I am very happy with the results I have gotten.

You can't beat this service for the price. You WILL have to go in an edit afterwards because the machine isn't by any means perfect, but if I provide a quality audio recording for it to work with, the editing time was minimal. I remove all music and just send pure vocal, which I imagine helps. I have a podcast and like to provide a time stamped transcript. It would be unaffordable for me to use a human transcriber, so I am thrilled with Temi!

Very valuable for high quality recordings. Pretty accurate and insanely cheap rate. For multiple voices and not so good quality it's not worth even the really cheap price. Use Rev.com when you have many voices or poor quality audio, but Temi.com is awesome if you have just one or two people on a good quality recording.

I have done several transcripts with Temi, it's very fast turnaround and the editing environment has some interesting functionality but it's not useful for multi-party conversations and only somewhat useful to capture one side of a conversation (e.g. a Q&A where Q and A are in separate channels). Still a very high error rate compared to human (e.g. Rev.com). I continue to experiment with it but cannot recommend it a this time.

I love Temi and consider it a secret weapon. I run a film production company and use it for almost every project. It allows me to quickly and easily move around material to do paper edits for our films. It also allows me to share raw material with clients, editors, writers, and other producers. It is extremely cost effective, especially in consideration of the immense amount of time it saves me on every project.

Super-cool idea, and way cheaper than person-based transcription, but def takes effort to review/edit the transcription (though less than fully transcribing the original audio). Would be better if there was account-specific machine learning, so that I could upload a reviewed document and have the system learn from its initial mistakes. (e.g., the acroynym NASDDDS kept coming out as "nasty's" -- that would be an easy fix).

a good many of the words were spelled completely wrong but it was still helpful as a starter and I will definitely use you again

The turnaround was great but quality of the transcription was unusable. I had to send the audio file to Rev for human transcription.

Really poor quality transcription. Didn't do a good job setting expectation that only very high quality audio would work.