aiOla, a provider of speech recognition technology, has launched Whisper-Medusa, an open-source AI model to improve the processing speed and latency of automatic speech recognition.
The model allows increased token prediction and operates by initially freezing primary components of the Whisper program while training additional parameters, using Whisper transcriptions as training modules for Medusa.
Additionally, the company claims that the model will reduce generation runtimes.
Israel-based aiOla develops AI-powered speech recognition technology to digitize, automate, and streamline business processes across various industries. The aiOla's proprietary platform combines voice recognition, natural language understanding (NLU), and domain-specific language models to capture and process spoken data.
By using this site, you agree to allow SPEEDA Edge and our partners to use cookies for analytics and personalization. Visit our privacy policy for more information about our data collection practices.