Introducing DeepVA’s latest innovations: Changelog September 2022 – May 2023

We are thrilled to introduce our latest changelog video, which highlights the recent advancements and improvements made to our state-of-the-art software DeepVA.

September Update: Introducing Speaker Dataset Creation

In September, DeepVA introduced an innovative mining module called Speaker Dataset Creation. This cutting-edge module allows users to automate the retrieval of audio-based datasets, similar to our existing Face Dataset Creation module. By extracting and saving audio files with the speaker’s name, training becomes even more streamlined and automated. This feature empowers media companies to harness the power of audio data effortlessly. 

December Update: Unleashing Speech Recognition & Speaker Identification Module

DeepVA’s December update brought a game-changing addition to our platform—a major audio mining module, Speech Recognition. This powerful module combines Speech-to-Text functionality with Speaker Identification, which makes it significantly easier to analyze vast amounts of audio data. Moreover, it automatically identifies the spoken language and offers translation capabilities, opening new doors for cross-lingual analysis. This module lays the foundation for future enhancements and functionalities, ensuring our users stay ahead of the curve.

Latest Release: Enhanced Face Recognition with Similar Identity Detection

Last month, we unveiled an exciting enhancement to our Face Recognition capabilities. DeepVA now offers an additional feature that returns the most similar identities. Media companies can leverage this feature for a variety of applications, from personalized content recommendations to engaging user experiences. In addition to the similarity detection feature, DeepVA’s Face Recognition function is also utilized to increase accuracy in scenarios where the objective is to determine whether a certain person truly appears in a video. By employing this function, users can obtain the top 5 predictions, allowing for more precise identification and verification tasks. With these advancements, DeepVA empowers users with enhanced face recognition capabilities, providing accurate results and expanding the range of potential applications.

Open API Swagger Specification: Empowering Developers

As part of our commitment to accessibility and flexibility, we are thrilled to announce that DeepVA has made the Open API Swagger specification available for all endpoints. This documentation empowers developers to seamlessly integrate our software into their existing workflows and systems. By providing easy access to our API specifications, we aim to foster collaboration and enable media companies to leverage the full potential of DeepVA.

Stay updated with the latest news and product releases by checking out our Changelog, subscribing to our newsletter or following us on LinkedIn.


latest AI news

Subscribe to our newsletter

Don’t worry, we reserve our newsletter for important news, so we only send a few updates once in a while. No spam!

🎉 Introducing Deep Live Hub: Your All-in-One Live Streaming Solution!  
This is default text for notification bar