fal-ai Speech-to-Text fal/speech-to-text Leverage the rapid processing capabilities of AI models to enable accurate and efficient real-time speech-to-text transcription. Input: speech Output: text RuntimeWire coverage Speko routes voice AI across providers using language-specific benchmarksSunbird relaunches iMessage on Android after 2023 security shutdownKevin Rose ships a free prompt for probing what AI remembers about youNous Research ships Hermes Agent with real-time voice and agent-to-agent linksCostPerPrompt publishes calculators for estimating monthly AI application costsSpeko launches a benchmark-based router for voice AI modelsNous Research adds local wake-word activation to Hermes AgentFish Audio raises $50 million seed for open-weight voice modelsTelnyx adds Moonshot's 2.8 trillion-parameter Kimi K3 to its inference APISpacebar unveils shared runtime for real-time multimodal AI agentsHandy creator released transcribe.cpp to unify local speech-to-text modelsGradium extends seed to $100 million as Nvidia backs voice AI infrastructure View on OpenRouter. Model data sourced from OpenRouter.