A speech-to-text model using GPT-4o for transcribing audio. It offers improved word error rate, better language recognition, and higher accuracy compared to the original Whisper models. Use it for more precise transcripts.
Not enough throughput data
Not enough TTFT data