Play.ht — AI Voice Generator
Low-latency TTS and voice cloning with 200+ realistic voices and a developer-friendly API.
Sorting
NVIDIA Riva
GPU-accelerated ASR/TTS SDK for low-latency, on-prem or cloud voice AI deployments.
Speechmatics Speech-to-Text
High-accuracy, low-latency enterprise ASR with multilingual/code-switching and real-time or batch modes.
GroqCloud
Ultra-low latency LPU-powered inference for text, speech, and vision models.
RunPod Serverless Endpoints
Always-on, pre-warmed GPU endpoints for low-latency model inference at scale.