We're hiring!
*

Transforming speech technology with WhisperLive

Kara Bembridge avatar

Kara Bembridge
May 28, 2024

Share this post:

Reading time:

The world of AI has made leaps and bounds from what it once was, but there are still some adjustments required for the optimal outcome. In the realm of conversational AI, VoxAI had already developed a platform to capture customer orders. The response time and oratory abilities needed improvement and this is where Collabora stepped in with WhisperLive.

WhisperLive, a real-time transcription service powered by OpenAI's Whisper model departed from traditional speech recognition methods by incorporating voice activity detection (VAD). VAD identifies speech presence, allowing for selective transmission of audio data to enhance transcription accuracy while optimizing data handling.

Simultaneously, Collabora employed a finely tuned Mistral model for the NLP component. Renowned for its efficiency and versatility, Mistral is six times faster and equally or more effective than the Llama 2 70B model across benchmarks. This model supports multiple languages and possesses inherent coding capabilities.

As we look to the WhisperLive's promising possibilities, our Machine Learning Lead, Marcus Edel, puts it best:

"The future of customer interaction lies in the harmonious fusion of sophisticated AI and powerful communication technologies. As we continue our mission and build fully in the open, WhisperLive, and now the award-nominated WhisperFusion, are poised to make an impact in the communication technology landscape."

To learn more about how this project came to life, take a look at out our case study.

If you're eager to implement your own transcription service, please get in touch! Our machine learning team is ready to assist you with your AI needs.

Comments (0)


Add a Comment






Allowed tags: <b><i><br>Add a new comment:


Search the newsroom

Latest Blog Posts

The latest on cmtp-responder, a permissively-licensed MTP responder implementation

12/06/2024

Part 3 of the cmtp-responder series with a focus on USB gadgets explores several new elements including a unified build environment with…

A roadmap for VirtIO Video on ChromeOS: part 3

06/06/2024

The final installment of a series explaining how Collabora is helping shape the video virtualization story for Chromebooks with a focus…

Hacking on the PipeWire GStreamer elements

05/06/2024

Last week I attended the GStreamer spring hackfest in Thessaloniki to work on the PipeWire GStreamer elements and connect with the community.

Transforming speech technology with WhisperLive

28/05/2024

The world of AI has made leaps and bounds from what It once was, but there are still some adjustments required for the optimal outcome.…

Re-converging control flow on NVIDIA GPUs - What went wrong, and how we fixed it

25/04/2024

While I managed to land support for two extensions, implementing control flow re-convergence in NVK did not go as planned. This is the story…

Automatic regression handling and reporting for the Linux Kernel

14/03/2024

In continuation with our series about Kernel Integration we'll go into more detail about how regression detection, processing, and tracking…

Open Since 2005 logo

We use cookies on this website to ensure that you get the best experience. By continuing to use this website you are consenting to the use of these cookies. To find out more please follow this link.

Collabora Ltd © 2005-2024. All rights reserved. Privacy Notice. Sitemap.