Full threadacossta·xAI says its new speech-to-speech model ranks #1 on Artificial Analysis's voice index, with under a second of latency at the same API priceView on HN