Hooho API enables developers to compare and implement
any private speech-to-text solutions (AWS, Google, Azure,...)
and train & deploy their own ASR models
Hooho API allows you to use different speech-to-text solution providers through a single implementation. You use the same query structure for your API calls, and your API respons are unifiedWe also allow you to deploy, train and use your own speech recognition model (ASR) in the cloud.
For the moment, we allow the use of speech recognition solutions from Google, AWS, Microsoft and Rev. We are constantly adding the solutions that we think are the most efficient. You can request a new integration by contacting us.
For each provider, we take care to implement all the available features. The goal is to allow you to take advantage of all the power of the solutions, but more easily, thanks to a single implementation.
We allow you to train and deploy in the cloud your own ASR application based on Pytorch models. You can easily train a model with your own data and compare the WER score to other solutions available on the market.
Based on how many requests you need to
send per month, we create a suitable plan.
If you need to deploy your own ASR application, we take into account the training and the number of model variants created. Contact us to get your estimate.
We maintain 99.99% uptime and are on call to respond to security alerts and events.
Hooho will compare your corrected script with the results
of each solution and return a WER score (Word Error Rate).
You can see which solution scores best for your project.
By changing a single parameter, you can send your request to different providers and,
if necessary, our documentation will generate the request
according to the desired solutions and configuration options.
