Voice & Multimodal
Vapi
Developer platform for voice AI agents over phone, web, and MCP
Category
Voice & Multimodal
Platforms that let agents perceive, generate, or interact through voice, audio, images, video, and other modalities.
Editorial scope version 1.0 · effective 3 September 2026
Category listing
1 Voice & Multimodal Tools
Definition
What belongs in Voice & Multimodal?
This category covers agent-first interaction layers where real-time speech or another non-text modality is central to the agent’s operation or user experience.
Includes real-time voice and other non-text agent interaction layers that manage modalities, transports, turn-taking, interruptions, state, media events, or handoffs.
Scope
Inclusion and exclusion criteria
Include when
- The product provides a documented runtime or interface for voice or multimodal agents.
- Real-time interaction, turn handling, telephony, media events, or modality coordination is substantive.
- Agents are active participants rather than simple media-generation callers.
Exclude when
- It is only a generic speech, image, or video model API.
- It adds voice or media to a conventional app without a material agent interaction model.
Selection guide
How to compare these tools
- Which modalities and transport channels are supported?
- How are latency, interruptions, state, and handoffs handled?
- What recording, consent, and data-retention controls exist?
Typical use cases
- Build agents that hold real-time voice conversations
- Connect agents to telephony or multimodal channels
- Coordinate interruptions, turn-taking, and human handoffs
- Manage consent, recordings, and retention for media interactions
Category evidence
Sources for this category guidance
- OpenAI GPT-Realtime-2 model documentation
The GPT-Realtime-2 documentation describes speech-to-speech agent workflows with text, image, and audio inputs, audio output, and tool use.
accessed 2026-09-03
Comparable facts
Voice & Multimodal comparison
Unknown values are shown explicitly rather than inferred from marketing copy. Open a profile to inspect its claim-level sources.
| Tool | Classification | Pricing | Interfaces | Deployment | Verification |
|---|---|---|---|---|---|
| Vapi | Agent-native | Paid | REST API, web dashboard, Web SDK, TypeScript server SDK, Python server SDK, command-line interface, MCP server, webhooks | hosted | documentation reviewed |