Scribewave - Fast & secure multilingual transcription

Hi@juan_smith1 , that's a great question, thank you! We constantly benchmark 14 different ASR models, both commercial and open-source. Some of them we fine-tuned ourselves on local dialects or smaller languages. This makes us confident that for any situation we can provide you with the best available transcript. That doesn't mean it's always perfect, but we're getting there. And you sure won't find it more accurate anywhere else!

@ethan_lennon yes, you can for example highlight parts in your transcript and export this as an edited video to Adobe Premiere Pro! For now, we don't offer a direct FCP integration, but I'll put it on the list of features-to-come. You can already export a subtitle file and import that in Final Cut though 😊

Hi @hina_siddiqui , yes: Scribewave provides the option to add custom vocabulary to your organization or individual files, which is perfect for the use case you describe. For enterprise clients, I can even offer a self-learning vocabulary that gets updated every time someone corrects a word in a transcript. In this way, you create your own, specialised TTS model.

@alex_bryant1 thanks for trying out the tool and sharing your experience! 🙏

@faith_foster a very important consideration indeed! Our models are trained to work even in these challenging circumstances. We can even automatically tag audio events such a laughter or music.

Hi @camden_granger , thank you for your kind words! Yes, the software can recognize and tag up to 30 speakers per file 🙂

@lucas_parsons Great question. Yes, it does! You can upload up to 100 files at the same time, and download the results in bulk as word or srt files 🚀

@mary_gen thank you for your question. We need to store your files on secure servers to make the editor work. But you stay owner of your data and upon project deletion, everything's removed. So if you upload your file, download the result and delete the project, you don't leave a trace on our servers. We don't use your data for training either.

Hi @ella_hall4 , thanks! We offer speaker diarization up to 30 speakers per file :)

Hi@katie_west2 , thank you for your feedback 😊 The tool currently supports only local file uploads, but adding support for cloud storage services is a great suggestion. I’ve looked into this integration before, and knowing it’s something users are actively interested in definitely pushes it higher on my priority list!

@luke_stanley1 real-time transcription for meetings is definitely something that I'm considering. For now, the platform only supports pre-recorded audio, but you can expect a meeting integration to appear in the near future.

@cady_gilmore you're completely right! Our models have been trained on lots of data, resulting in a remarkable accuracy on these languages. Feel free to give it a try, and let me know your feedback 😊