Gladia Unveils Solaria: A Multilingual Speech-to-Text Model

Gladia has launched Solaria, a next-generation speech-to-text model designed to support global scalability, announced in a press release. Solaria offers unmatched accuracy and speed, providing native-level transcription in 100 languages, including 42 that were previously underserved by other models.

The model is tailored for real-time communications in call centers and voice-first platforms, achieving a 94% Word Accuracy Rate in languages such as English, Spanish, and French. Solaria's ultra-low latency of 270 ms ensures natural and responsive conversations, making it ideal for enterprises seeking to enhance customer service operations with AI-powered voice technology.

Solaria's capabilities include native-level transcription, real-time code-switching, and translation across all supported languages. It also adapts to industry-specific terminology and extracts critical data, ensuring high accuracy even in noisy environments. With enterprise-grade data security, Solaria complies with GDPR, HIPAA, and SOC 2 standards.

Gladia's partnership with LiveKit further enhances Solaria's offerings by integrating real-time multilingual translation within AI-driven applications, providing developers with global language capabilities through seamless API integration.