Whisper Large v3 Turbo (speech-to-text)

Whisper Large v3 Turbo debuts with automatic speech recognition

Whisper Large v3 Turbo has been added to the OVH AI Endpoints catalog as a new speech-to-text model hosted in the GRA region. This is OpenAI's Whisper model optimized for faster inference while maintaining the capabilities of the large v3 architecture. The model provides automatic speech recognition functionality, converting spoken audio into written text across multiple languages. As a turbo variant, it is designed to offer reduced latency compared to the standard Whisper Large v3, making it suitable for applications requiring quicker transcription turnaround. The model handles various audio formats and can process speech in diverse acoustic conditions. Users gain access to state-of-the-art ASR technology through OVH's infrastructure in the Gravelines datacenter. This addition expands the available speech processing options for developers building voice-enabled applications, transcription services, or accessibility tools. The model's performance characteristics on text-intelligence benchmarks are not applicable given its audio-to-text nature, but it represents a significant capability addition to the platform's speech processing portfolio.