# Neuphonic

**Canonical:** https://apis.io/providers/neuphonic/  
**Website:** https://www.neuphonic.com/  
**APIs profiled:** 7

Neuphonic is an ultra-low-latency voice AI platform specializing in real-time text-to-speech synthesis with sub-25ms latency, making it suitable for conversational AI and live applications. The platform provides both a cloud-hosted API with WebSocket streaming and Server-Sent Events (SSE), as well as open-source on-device models (NeuTTS Air, NeuTTS Nano) that run without a GPU. Neuphonic supports nine languages including English, Spanish, German, French, Urdu, Japanese, Korean, Chinese, and Portuguese, and offers instant voice cloning from short audio samples. Developers can also build conversational AI agents via the Agent API, which integrates with GPT-4o and supports Model Context Protocol (MCP) servers. Authentication uses API keys passed via the X-API-KEY header for SSE and as a query parameter for WebSocket connections.

## Kin Score — 48.3 / 100 (developing)

Scored 2026-08-20 under rubric 0.12.0. Trend: flat (+0.0 from 48.3).

| Facet | Score |
|---|---|
| Discoverability | 74.1 |
| Contract Quality | 74.1 |
| Governance | 25.0 |
| Contract Governance | 25.0 |
| Operational Transparency | 50.0 |
| Developer Ergonomics | 25.0 |
| Commercial Clarity | 39.5 |
| Access Clarity | 39.5 |

## Agent readiness — 34.2 (agent-ready)

| Dimension | Value |
|---|---|
| Spec Presence | yes |
| Agentic Access | derived |
| Reversibility Documented | no |
| MCP Server | no |
| Auth Clarity | yes |
| Idempotency | no |
| Error Semantics | no |
| OpenAPI Examples | no |
| Rate Limit Signal | documented |
| Event Surface Described | yes |
| Agent Skills | no |
| Well Known Catalog | no |
| Consent Identity | no |
| Agent Card | no |
| Dry Run Mode | no |

## Access

Freemium · Self-serve signup — onboarding: self-serve, pricing: freemium, trial: no (confidence: high).

## APIs (7)

- **Neuphonic TTS WebSocket API** — WebSocket endpoint for continuous, low-latency text-to-speech streaming. Enables real-time voice synthesis with sub-25ms latency, supporting multiple text chunks over a single p...
- **Neuphonic Voice Cloning API** — REST API for creating and managing cloned voices. Accepts audio samples (MP3 or WAV, minimum 6 seconds, under 10MB) and generates a custom voice model. Supports creating, retrie...
- **Neuphonic Agent API** — REST API for creating and managing conversational AI voice agents. Agents combine Neuphonic TTS with GPT-4o for interactive voice applications and support Model Context Protocol...
- **Neuphonic Agents API** — The Agents API from Neuphonic — 3 operation(s) for agents.
- **Neuphonic Ping API** — The Ping API from Neuphonic — 1 operation(s) for ping.
- **Neuphonic Sse API** — The Sse API from Neuphonic — 2 operation(s) for sse.
- **Neuphonic Voices API** — The Voices API from Neuphonic — 2 operation(s) for voices.

## Agentic access (1)

- **Neuphonic Agentic Access** — 12 operations · 8 acting

## Security (2)

- **Neuphonic Authentication** — apiKey · 2 schemes
- **Neuphonic Domain Security** — TLSv1.3 · HSTS · DMARC

## Plans (1)

- **Neuphonic Plans Pricing**

## Tags

Text-to-Speech, Voice AI, Audio, Streaming, WebSocket, Voice Cloning, Conversational AI, Real-Time, Multilingual, On-Device AI

---

Profiled by [API Evangelist](https://apievangelist.com) and published on [APIs.io](https://apis.io/providers/neuphonic/). Scores are computed from the provider's own public artifacts under a published rubric.