Vatis Tech is a fast, accurate and insightful transcript and summary service. By offering an editable transcript, extracting a list of questions, mentioned entities and providing a short description based on most meaningful ideas, VatisTech saves countless hours when working with audio-video content.

Hi, Product Hunt! We are Vatis Tech (@dorumusuroi, @alexandru_topala, @einazare) and we’re very excited to launch here! During the past two months, we’ve been working on a product that saves you countless time when you transcribe or summarize audio content. We started Vatis Tech with the desire to index audio-video content in order to make it easy to search for information in it. But we thought that, while this on its own does bring a lot of value and we managed to achieve it by providing a transcript, we could go a step further and provide our users some pieces of information even before looking for them. Hence we extract the questions in the audio-video file, the mentioned entities such as people names, institutions, places or organisations, and we also try to provide a short description of the whole audio content by extracting the most meaningful ideas. We put all this together in what we call a summary. Currently we support 10 languages for transcript (English, Romanian, Czech, Danish, German, Spanish, Finnish, French, Italian,Swedish) and 5 languages for summary (English, Danish, French, Spanish, Italian). But they are continuously under development and this week will come with other new languages based on what our users request. We trust our product will save countless hours spent on transcribing and creating content notes to podcasters, journalists and all of you cool kittens that work heavily with audio and video. We’re happy to hear any feedback and questions you might have. By the way, if you register today you will get the first 90 minutes free. Thank you for giving it a try!

@dorumusuroi @alexandru_topala @einazare @adrianispas what's the underlying NLP tech/library that you're using to do the transcription and extraction?

Hi @chrismessina Thanks for your question! At the moment we're working closely with Google Cloud API tunned by our neural network for transcription. For summary, we run our neural networks to find the most relevant sentences from the audio content.

@adrianispas nice — def like the question-extraction feature!

@dorumusuroi @alexandru_topala @einazare @adrianispas Congratulations on the launch! Can it be used to address this need gap - 'Automatic transcript for podcasts with search' posted on my platform - https://needgap.com/problems/46-... ?

Hi, @needgap! Thanks for your support and your question! At the moment in our app you can upload your podcast episode, in a few moments your transcript it's ready. The transcript is searchable and editable, and you can export your transcript as captions (SRT file) for your video.

Cool tool! Especially with the forced digitisation going around. I'll try it out for some team meetings, I think it would be cool to extract or look for exact keywords / topics after a long call. Everybody has their use-case!

@sk_dub Hi Alin and thank you! This is definitely a use case we had in mind when we built the platform and we hope it's going to serve you well. We definitely want to keep improving on it so do let us know of any feedback you will have.

Great product! Well done!