Tagged: speech-to-text

10 articles

Latest

Grok Voice 2.0 cuts word errors by 67% for short phrases at $0.10/hr

Voice Agents Demand Sub-Second Latency, Not Just Better LLMs

Why Your Voice AI Feels Slow Even When Your LLM's TTFT Looks Fast

Google's Gemini 3.5 Transcribe hits 2.6% error rate but splits into two incompatible APIs

Google's Gemini 3.5 Transcribe hits 4% error rate across 85 languages in real time

Google's Gemini 3.5 Transcribe cuts voice-to-text lag by 70% and cleans up your 'ums'

Superwhisper open-sources a 0.6B text normalizer that cleans up messy transcripts

Whisper transcribes 2x faster with speculative decoding, zero accuracy loss

OpenAI Slashes GPT-3.5 API Costs by 90%, Whisper Goes Public

OpenAI's new audio models can finally take direction