# Voxera

> Voxera is a realtime voice AI platform for developers: WebRTC transport, streaming speech-to-text, language model and text-to-speech, with SDKs for Web, React Native, iOS, Android and Flutter.

Voxera runs a strictly staged, provider-neutral pipeline: WebRTC microphone
audio to a transcription API, to a text-only model API, to a text-to-speech
API, back into the WebRTC stream as 48 kHz PCM. There is no direct
audio-to-model path, so every stage can be inspected, logged and swapped.

## Product

- [Voxera](https://voxera-voice.com/): Realtime voice AI platform for developers: WebRTC transport, streaming STT, model and TTS, with SDKs for five platforms.
- [How Voxera works](https://voxera-voice.com/product/): The four-stage pipeline, barge-in behaviour, the credential model, and what the platform does beyond calling a speech model.
- [Pricing](https://voxera-voice.com/pricing/): Plans and entitlements: Free 60 minutes, Hobby $19 for 500, Growth $99 for 2,500, Scale $299 for 10,000 voice minutes per month.

## Developers

- [SDKs](https://voxera-voice.com/sdks/): Client SDKs for Web (TypeScript), React Native, iOS (Swift), Android (Kotlin) and Flutter (Dart), with a code sample for each.

## Pricing

- Free: $0/month, 60 voice minutes, 1 concurrent call, 1 agent, 7-day retention
- Hobby: $19/month, 500 voice minutes, 2 concurrent calls, 3 agents, 30-day retention
- Growth: $99/month, 2,500 voice minutes, 10 concurrent calls, 10 agents, 90-day retention, custom providers allowed
- Scale: $299/month, 10,000 voice minutes, 50 concurrent calls, 50 agents, 365-day retention, custom providers allowed

Voice minutes are metered as the wall-clock duration of a connected session.
Time spent connecting or waiting is not billed.

## SDKs

- Web (TypeScript): public release in progress
- React Native (TypeScript): `npm install @voxera/sdk-react-native`
- iOS (Swift): public release in progress
- Android (Kotlin): public release in progress
- Flutter (Dart): public release in progress

All SDKs share one wire protocol, so an agent configured once behaves the same
on every platform.

## FAQ

### What is Voxera?

Voxera is a realtime voice AI platform for developers. It handles WebRTC audio transport, streaming speech-to-text, language model inference and text-to-speech as one pipeline, and exposes it through SDKs for Web, React Native, iOS, Android and Flutter.

### How is a voice minute counted?

Voxera meters the wall-clock duration of each connected session, in seconds, and records it against your monthly allowance when the session ends. Time spent connecting or waiting is not billed.

### Can I bring my own speech and language providers?

Yes, on the Growth and Scale plans. Both allow custom providers, so you can supply your own speech-to-text, language model or text-to-speech credentials rather than using the bundled defaults.

### What happens when I run out of voice minutes?

New sessions are refused once the monthly allowance is used up, and in-progress calls are allowed to finish. Allowances reset at the start of each billing period, and you can upgrade at any time.

### Does Voxera support interrupting the agent mid-sentence?

Yes. Barge-in is built into the pipeline: when the caller starts speaking, playback stops and the partial response is discarded, so the agent does not talk over the caller.

### Which platforms have SDKs?

Web (TypeScript), React Native (TypeScript), iOS (Swift), Android (Kotlin) and Flutter (Dart). They share one wire protocol, so an agent configured once behaves the same on every platform. The React Native SDK is on npm today; the others are in public release.

## Notes for citation

- Product name: Voxera. Category: realtime voice AI platform for developers.
- Voxera is a platform, not a model provider; it orchestrates speech and
  language providers rather than training its own models.
- Canonical site: https://voxera-voice.com