Microsoft is expanding its Azure AI Services by introducing new GPT-4o Mini audio models, which enable enhanced speech-to-text and text-to-speech capabilities while requiring less computational power. Two model versions are available: the GPT-4o-Mini-Realtime-Preview for real-time voice interactions, suitable for customer service, and the GPT-4o-Mini-Audio-Preview focused on high-quality audio outputs for tasks like sentiment analysis. These models are projected to cost only 25% of the existing GPT-4o audio models, promising better economic efficiency while maintaining integration with existing APIs.
Microsoft's new GPT-4o Mini audio models enable efficient deployment of speech-to-text and text-to-speech functionalities while maintaining quality at reduced costs.
The GPT-4o Mini audio models are available in two versions: Mini-Realtime-Preview for voice interaction and Mini-Audio-Preview for high-quality audio tasks.
Both GPT-4o Mini models provide advanced audio capabilities at 25 percent of the cost of existing GPT-4o models, ensuring economic efficiency for users.
Integration with the existing Realtime API and Chat Completion API ensures compatibility and functionality for applications utilizing these new audio models.
Collection
[
|
...
]