Whisper Large v3 Turbo (speech-to-text)

Whisper Large v3 Turbo enters catalog with ASR capabilities

Whisper Large v3 Turbo by OVH AI Endpoints has been newly added to the benchmark catalog with automatic speech recognition capabilities. This speech-to-text model represents OpenAI's Whisper architecture optimized for faster inference while maintaining transcription quality. As a new entry, there is no performance history to compare against previous benchmark windows. The model is designed to convert spoken audio into written text across multiple languages, serving use cases such as transcription services, voice command processing, and accessibility applications. Users should note that this is an audio input model focused specifically on ASR tasks, not a general-purpose language model. The turbo variant indicates optimization for reduced latency compared to the standard Whisper Large v3, making it suitable for applications requiring faster processing times. As this model processes audio rather than generating text responses, it falls outside the scope of Tokonomix text-intelligence benchmarks. Future verdicts will track any changes in capability coverage or service availability as the model continues operation within the OVH AI Endpoints infrastructure.