AssemblyAI

API for customizable speech recognition

get it

Reviews

Discussion

Hunter
Makers
You need to become a Contributor to join the discussion - Find out how.
Dylan FoxMaker@youvegotfox · AssemblyAI
Hey all, I’m the founder of AssemblyAI (https://www.assemblyai.com). We're in the current batch of Y Combinator (S17) and are building an API for customizable speech recognition. Developers and companies use our API for things like transcribing phone calls and building voice powered smart devices. Unlike current speech recognition APIs, developers can customize our API to more accurately recognize an unlimited amount of industry specific words or phrases unique to what they're building without any training required. For example, you can recognize thousands of product or person names with our API. Or you can more accurately recognize commands/phrases common or custom to your use case. We've developed our own deep neural network speech recognition architecture, and aren't using any open source speech frameworks like Kaldi or Sphinx (just Tensorflow). Because of this, we're able to run things more affordably and pass those savings on to developers. I used to work on projects that had speech recognition requirements before starting AssemblyAI, and saw how limiting, expensive, and hard to work with traditional speech recognition services and APIs were. We want to help developers and companies easily build products with speech recognition. Would love feedback from the community on what we're building, and if you have any questions about deep learning or deep learning in production ask away!
Moin UddinPro@moinism · Creator/Developer
@youvegotfox congrats on the launch. Are you guys working on a JS SDK for browser use? Would help you I think.
Dylan FoxMaker@youvegotfox · AssemblyAI
@moinism hey thanks! we are definitely working on a JS SDK that uses WebRTC ... hoping to have it out soon!
Nearly Normal@nearlynormal · If it goes without saying, let it.
@youvegotfox Great going Dylan. Look forward to libraries in JS and PHP!
Daniel Williams@xcadaverx
@youvegotfox hi! This seems really neat. Is this a potential competitor for specialized recognition services, such as medical niches? Could you for example, have one user be more specialized in radiology lingo while another user is more specialized for cardiology? Does it learn over time with edits from the user and such? I work on an app that uses a 3rd party speech recognition SDK in the medical field, and I'm always looking for other options.
Dylan FoxMaker@youvegotfox · AssemblyAI
@xcadaverx Exactly! You can have one user for radiology lingo and another for cardiology. And when you query the API with audio data, the transcripts will be customized for each user. Right now there is no user feedback loop, but we do QA that improves the system over time with more use. Would love to chat with you about your app! If you want to email beta@assemblyai.com I will look for your email and follow up with you directly about trying out our API!
Sotiris Karagiannis@skaragiannis
Seems very interesting product. I'd love to test it. Already shared it with some friends :-)
Dylan FoxMaker@youvegotfox · AssemblyAI
@skaragiannis thanks! if you email beta@assemblyai.com I'll look for your mail and can get you setup with the API
Chris Wilson@automateiq · VoiceCut CEO
@youvegotfox We built VoiceCutTech.com and might look into using your service. How are you different than VoiceBase and are you better at recognizing custom commands? What is or will be your pricing model? Interested to find out more
Dylan FoxMaker@youvegotfox · AssemblyAI
@automateiq Hey Chris! If you want to shoot an email to beta@assemblyai.com and mention this post, I can look for it and reply to you with some more info. We're very accurate at recognizing custom commands. You can actually restrict the API to just supporting the commands you need to be able to support.