NVIDIA Riva Studio Description
The browser has in-app prompts that allow you to record and use the recording tool. To create a 30-minute dataset to train a TTS model, a predefined set of phonetically-balanced sentences is available. You can make the model sound just like you by selecting the range that best suits your voice pitch. The standard voice pitch setting for human voices is already included. You can also use the preprogrammed best recipes to personalize the TTS model for you voice. To integrate a customized TTS model in your application, generate an API. Download a deployable package that includes a helm chart for use on any cloud or on-premises Kubernetes cluster. You can then automatically host your voice microservice using NVIDIA or set it up with only one line of code. Riva TTS models are easy to set up, customize, deploy, and manage with no infrastructure configuration.
Company Details
Media
