Google debuts Gemini 3.1 Flash Live audio

Google is closing the "vibes" gap with OpenAI's Advanced Voice Mode while offering superior developer access via a native API. By prioritizing low-latency multimodal streaming, they are positioning Gemini as the foundation for a new wave of voice-native applications. Continuous video and screen stream processing gives it a major edge over GPT-4o's snapshot-based vision for live tasks, while improved tonal awareness makes interactions feel significantly more human and intuitive. A 90.8% score on audio-based function calling benchmarks suggests high reliability for complex workflows, supported by integrated SynthID watermarking and competitive pricing for the Flash tier.