# rumik silk tts

> text-to-speech api by rumik ai.

## Docs

- [overview](https://docs.rumik.ai/index.md): the rumik silk text-to-speech api.
- [quickstart](https://docs.rumik.ai/quickstart.md): synthesize your first clip in three steps.
- [pay as you go](https://docs.rumik.ai/pricing/pay-as-you-go.md): use the silk services you need and pay for your usage.
- [unlimited](https://docs.rumik.ai/pricing/unlimited.md): unlimited generation with a concurrency capacity that fits your needs.
- [silk muga 1](https://docs.rumik.ai/muga.md): our expressive text-to-speech voice.
- [silk mulberry 1.5](https://docs.rumik.ai/mulberry.md): our faster, description-driven text-to-speech voice.
- [prompting silk muga 1](https://docs.rumik.ai/prompting-muga.md): how to steer silk muga 1 with tones and inline events.
- [prompting silk mulberry 1.5](https://docs.rumik.ai/prompting-mulberry.md): how to steer silk mulberry 1.5 with a natural-language description.
- [numbers, ids & dates](https://docs.rumik.ai/normalization.md): how to write text so silk speaks numbers, money, dates, and ids correctly.
- [stream in real time](https://docs.rumik.ai/streaming.md): low-latency playback over a websocket session.
- [python sdk](https://docs.rumik.ai/sdk/introduction.md): the official rumik-ai python package for silk text-to-speech.
- [quickstart](https://docs.rumik.ai/sdk/quickstart.md): install the sdk, authenticate, and synthesize your first clip.
- [synthesis](https://docs.rumik.ai/sdk/synthesis.md): batch, streaming, sessions, and async with the rumik-ai client.
- [errors and configuration](https://docs.rumik.ai/sdk/errors.md): the exception tree, retries, timeouts, headers, and debug logging.
- [sdk reference](https://docs.rumik.ai/sdk/api-reference.md): every public class, method, exception, and type in rumik-ai.
- [build a voice agent with pipecat](https://docs.rumik.ai/pipecat.md): drop rumik into a pipecat pipeline with pipecat-rumik.
- [build a voice agent with livekit](https://docs.rumik.ai/livekit.md): drop rumik into a livekit agent with livekit-plugins-rumik-ai.
- [coding-agent skill](https://docs.rumik.ai/agent-skill.md): hand your ai coding agent everything it needs to integrate rumik tts correctly.
- [cookbook](https://docs.rumik.ai/cookbook/index.md): step-by-step guides for building with rumik tts.
- [build a voice agent with pipecat](https://docs.rumik.ai/cookbook/voice-agent-pipecat.md): a complete, from-scratch walkthrough: a real-time voice agent that talks back in hinglish.
- [build a voice agent with livekit](https://docs.rumik.ai/cookbook/voice-agent-livekit.md): a complete, from-scratch walkthrough: a real-time livekit voice agent that talks back in hinglish.
- [pipecat pipeline over livekit transport](https://docs.rumik.ai/cookbook/voice-agent-pipecat-livekit.md): use livekit purely as the webrtc transport while pipecat runs the stt to llm to rumik tts pipeline.
- [synthesize speech](https://docs.rumik.ai/api-reference/speech/synthesize-speech.md): synthesize an utterance and return it as a 24 khz mono wav file. pass your api key as a bearer token. payg tts is prorated from the exact submitted `text` character count, including whitespace, punctuation, newlines, and inline tags. generated audio duration is measured server-side for telemetry onl…
- [create a streaming session](https://docs.rumik.ai/api-reference/speech/create-a-streaming-session.md): mint a one-shot websocket session for real-time streaming synthesis. returns a `ws_url` and a `token`. connect to `ws_url?token=<token>`, send a json synthesis frame, then receive raw pcm int16 little-endian @ 24 khz mono binary frames followed by a terminal `{"type":"done"}` JSON frame.

## OpenAPI Specs

- [openapi](https://docs.rumik.ai/openapi.json)