Cartesia
Speech generation, transcription and managed tools for voice agents
Speko
Route voice AI requests across speech models and compare language benchmarks
Meet Speko
Speko provides infrastructure for choosing and connecting the models behind voice applications. Its hosted Router offers a common interface for speech recognition, language models, and speech synthesis. Developers can compare benchmark coverage by language and examine quality, latency, and cost measurements when deciding which supported model fits a particular stage of the conversation.
A separate Gateway runtime integrates with LiveKit and Pipecat, supporting direct provider streaming, locally managed credentials, and optional Speko-managed routes. Public API contracts and an MCP interface offer additional ways to configure integrations. Benchmark availability differs across models and languages, so the tables are a starting point for evaluation rather than proof of performance on every application's audio.
Discussion
Tried it? Share a useful thought or ask a question.
Good conversations start with a hello.
Sign in to commentProduct details
- Website
- speko.ai
- Status
- Launched
- Category
- Artificial intelligence
- Listed
Explore related topics
Similar products
Fish Audio
Create expressive speech, voiceovers, and conversational audio
Kokori
Generate speech locally on macOS through a desktop app and REST API.
Smallest.ai
Turn text into speech for narration and real-time applications
ElevenLabs
Create voice and audio with AI.
Murf
AI voiceovers, speech APIs, and conversational agents