# MiniMax

> Text-to-speech with MiniMax's T2A API: MiniMaxHttpTTSService streams PCM audio over HTTP.

Source: https://polvallverdu.github.io/pipedog/docs/services/tts/minimax/



## Overview [#overview]

`MiniMaxHttpTTSService` posts each sentence to MiniMax's T2A (Text-to-Audio) endpoint with streaming on, and pushes the hex-encoded PCM audio of each streamed `data:` block as it arrives.

<Cards>
  <Card title="Upstream page" href="https://docs.pipecat.ai/api-reference/server/services/tts/minimax">
    Pipecat's documentation for the same service.
  </Card>

  <Card title="Source" href="https://github.com/polvallverdu/pipedog/tree/main/packages/minimax/src">
    `packages/minimax/src/tts.ts`.
  </Card>

  <Card title="MiniMax docs" href="https://platform.minimax.io/docs/api-reference/speech-t2a-http">
    The T2A HTTP API reference.
  </Card>
</Cards>

## Installation [#installation]

```bash
bun add @pipedog/minimax
```

```ts
import { MiniMaxHttpTTSService } from "@pipedog/minimax/tts";
```

## Prerequisites [#prerequisites]

* A MiniMax account, an API key and your group id.
* Pass them as `apiKey` and `groupId`. The examples read them from `MINIMAX_API_KEY` and `MINIMAX_GROUP_ID`.

## Configuration [#configuration]

| Option | Type | Required | Description |
|---|---|---|---|
| `sampleRate` | `number \| null \| undefined` | no | Output sample rate for generated audio. |
| `skipTextTypes` | `string[] \| null \| undefined` | no | List of text types that should not be spoken. |
| `textTransforms` | `[string, TextTransform][] \| null \| undefined` | no | A list of TextTransforms to transform text just before sending it to TTS. Each TextTransform takes the aggregated text and its type, and returns the transformed text. To register, provide a list of tuples of `[textType \| "*", transformFunction]`. |
| `transportDestination` | `string \| null \| undefined` | no | Destination for generated audio frames. |
| `apiKey` | `string` | yes | MiniMax API key for authentication. |
| `baseUrl` | `string \| undefined` | no | API base URL, defaults to MiniMax's T2A endpoint. - Global: https://api.minimax.io/v1/t2a_v2 - Mainland China: https://api.minimaxi.chat/v1/t2a_v2 - Western United States: https://api-uw.minimax.io/v1/t2a_v2 |
| `groupId` | `string` | yes | MiniMax Group ID to identify project. |
| `fetch` | `{ (input: RequestInfo \| URL, init?: RequestInit): Promise<Response>; (input: string \| URL \| Request, init?: RequestInit): Promise<Response>; } \| null \| undefined` | no | `fetch` used for HTTP requests. Defaults to the global `fetch`. DIVERGENCE: upstream takes a required shared `aiohttp_session`. |
| `stream` | `boolean \| undefined` | no | Whether to use streaming mode. Defaults to true. |
| `settings` | `MiniMaxTTSSettings \| null \| undefined` | no | Runtime-updatable settings. |
| `name` | `string \| null \| undefined` | no | Custom name; defaults to `ClassName#count`. |
| `taskManager` | `BaseTaskManager \| null \| undefined` | no | Task manager for handling tasks. |
| `enableDirectMode` | `boolean \| undefined` | no | Process frames immediately instead of using internal queues. |
| `metrics` | `FrameProcessorMetrics \| null \| undefined` | no | Optional metrics collector for this processor. |
| `textAggregationMode` | `TextAggregationMode \| null \| undefined` | no | How to aggregate incoming text before synthesis. `TextAggregationMode.SENTENCE` (default) buffers until sentence boundaries, `TextAggregationMode.TOKEN` streams tokens directly for lower latency. |
| `pushTextFrames` | `boolean \| undefined` | no | Whether to push TextFrames and LLMFullResponseEndFrames. |
| `pushStopFrames` | `boolean \| undefined` | no | Whether to automatically push TTSStoppedFrames. |
| `pushStartFrame` | `boolean \| undefined` | no | Whether to automatically create audio contexts and push TTSStartedFrames. When true, the base class handles `createAudioContext` and yields `TTSStartedFrame` before each synthesis call, so `runTts` implementations do not need to. |
| `stopFrameTimeoutS` | `number \| undefined` | no | Idle time before pushing TTSStoppedFrame when pushStopFrames is true. |
| `pushSilenceAfterStop` | `boolean \| undefined` | no | Whether to push silence audio after TTSStoppedFrame. |
| `silenceTimeS` | `number \| undefined` | no | Duration of silence to push when pushSilenceAfterStop is true. |
| `pauseFrameProcessing` | `boolean \| undefined` | no | Whether to hold incoming frames after a turn until its audio contexts end. Needed by services whose audio doesn't carry its request's context ID. |
| `maxConsecutiveZeroAudioContexts` | `number \| undefined` | no | How many consecutive TTS contexts may complete without producing any audio before the service is reported unable to do its job. Catches a provider that accepts requests and stays silent — an unknown voice ID, say — which no error ever surfaces. Every silent context reports an error the service can carry on from, whatever the limit is; on reaching the limit the service reports a permanent error instead, stops being given work, and the pipeline worker applies its `ProcessorUnusablePolicy`. Set to 0 to report silent contexts without ever writing the service off. |
| `appendTrailingSpace` | `boolean \| undefined` | no | Whether to append a trailing space to text before sending to TTS. This helps prevent some TTS services from vocalizing trailing punctuation (e.g., "dot"). Only applied in sentence aggregation mode; when streaming tokens, the incoming text's own whitespace is preserved. |
| `reuseContextIdWithinTurn` | `boolean \| undefined` | no | Whether the service should reuse context IDs within the same turn. |
| `audioCache` | `TTSAudioCacheStore \| null \| undefined` | no | Where to keep synthesized audio for reuse (e.g. a shared `TTSAudioCache`). Text whose audio is in the cache is served from it, with no request to the provider. Lines get there through `warmAudioCache()`, and also from live synthesis when `cacheSynthesizedAudio` allows it. Only services that yield their audio from `runTts` (HTTP and streaming-RPC services, see `supportsAudioCache`) use it, and only in sentence aggregation mode. Defaults to null (no cache). pipedog only. |
| `cacheSynthesizedAudio` | `boolean \| ((text: string) => boolean) \| undefined` | no | Whether audio synthesized live is stored in `audioCache`, or a function that decides per text. Defaults to false: only warmed lines are stored, since a generated reply can repeat what the user said. pipedog only. |
| `audioCacheNamespace` | `string \| undefined` | no | Extra part of the cache key, for constructor options that change the audio but are not in `settings`. pipedog only. |

### Settings [#settings]

Runtime-updatable settings, passed as `settings: new MiniMaxHttpTTSService.Settings({...})`
(`MiniMaxTTSSettings`). Any setting you leave out keeps the default below.

| Setting             | Type               | Default              | Description                                                                                       |
| ------------------- | ------------------ | -------------------- | ------------------------------------------------------------------------------------------------- |
| `model`             | `string`           | `"speech-2.8-turbo"` | Speech model.                                                                                     |
| `voice`             | `string`           | `"Calm_Woman"`       | Voice id.                                                                                         |
| `language`          | `Language \| null` | `null`               | Not sent. Use `languageBoost`.                                                                    |
| `speed`             | `number \| null`   | `1.0`                | Speed, 0.5 to 2.0.                                                                                |
| `volume`            | `number \| null`   | `1.0`                | Volume, above 0 up to 10.                                                                         |
| `pitch`             | `number \| null`   | `0`                  | Pitch, -12 to 12.                                                                                 |
| `emotion`           | `string \| null`   | `null`               | Emotional tone: `happy`, `sad`, `angry`, `fearful`, `disgusted`, `surprised`, `calm` or `fluent`. |
| `textNormalization` | `boolean \| null`  | `null`               | Text normalization (Chinese and English).                                                         |
| `latexRead`         | `boolean \| null`  | `null`               | Read LaTeX formulas.                                                                              |
| `languageBoost`     | `string \| null`   | `null`               | A language name to boost, such as `English`. `languageToMinimaxLanguage()` maps a `Language`.     |

## Usage [#usage]

```ts title="tts.ts"
import { TTSUpdateSettingsFrame } from "@pipedog/core/frames/frames";
import { MiniMaxHttpTTSService } from "@pipedog/minimax/tts";

const tts = new MiniMaxHttpTTSService({
  apiKey: process.env.MINIMAX_API_KEY ?? "",
  groupId: process.env.MINIMAX_GROUP_ID ?? "",
  settings: new MiniMaxHttpTTSService.Settings({
    model: "speech-2.8-turbo",
    voice: "Calm_Woman",
    emotion: "calm",
  }),
});
```

Change settings mid-call with a `TTSUpdateSettingsFrame`:

```ts
// Queue it on the worker, or push it from a processor upstream of the service.
const update = new TTSUpdateSettingsFrame({
  delta: new MiniMaxHttpTTSService.Settings({ speed: 1.2 }),
});
```

## Notes [#notes]

* `baseUrl` picks the region: `https://api.minimax.io/v1/t2a_v2` (global, the default), `https://api.minimaxi.chat/v1/t2a_v2` (mainland China) or `https://api-uw.minimax.io/v1/t2a_v2` (western United States). The group id is added as `?GroupId=`.
* The last streamed block repeats the whole utterance and carries `extra_info`. It is skipped, so no audio is played twice.
* The audio is 16-bit PCM at the pipeline's sample rate, one channel.
* `MiniMaxTTSSettings.fromMapping()` also accepts the older nested `voice_setting` dict (with `vol` for `volume`).

## Runtime notes [#runtime-notes]

* It uses the global `fetch`, so it works on Bun, Node, Deno and Cloudflare Workers. Pass `fetch` to use another one (upstream takes an aiohttp session).
* The deprecated `params` (`InputParams`), `model` and `voiceId` constructor options are not ported. Use `settings`.
* Tested against a local server, not against MiniMax itself.

| Runtime            | Works | What it needs |
| ------------------ | ----- | ------------- |
| Bun                | ✅ yes | Nothing.      |
| Node.js            | ✅ yes | Nothing.      |
| Deno               | ✅ yes | Nothing.      |
| Cloudflare Workers | ✅ yes | Nothing.      |

Generated from `pipecat.runtimes` and `pipecat.compat` in `packages/minimax/package.json`: ✅ means the compat smoke test passes on that runtime.