> **Building with AI coding agents?** Install the authstack plugin with one command. This equips your agent with accurate Scalekit implementation patterns.
>
> **Recommended** (the `-y` flags skip prompts, so the command runs without an interactive terminal):
> ```bash
> npx -y @scalekit-inc/cli setup -y
> ```
>
> Global:
> ```bash
> npm install -g @scalekit-inc/cli
> scalekit setup -y
> ```
>
> Supports Claude Code, Cursor, GitHub Copilot, Codex + skills for other Agent Skills-compatible agents in `~/.agents/skills`.
> Skills: integrate-agentkit, implement-saaskit, add-mcp-oauth, implement-sso, implement-scim.
> [Full setup guide](https://docs.scalekit.com/dev-kit/build-with-ai/)

---

# Connect AI agents to the Cartesia MCP server

The Cartesia MCP connector routes your AI agent's tool calls to Cartesia's own MCP server through Scalekit. Each user connects their own Cartesia access token once, and Scalekit sends it with every call, so your agent never handles credentials. It comes with 19 tools.

**Authentication:** Bearer Token
**Categories:** AI, Media
**Tools:** 19: 9 read, 7 write, 3 destructive
**Users sign in with:** Access token
**Built by:** Vendor MCP
**Try it:** [Playground in the Scalekit dashboard](https://app.scalekit.com/ws/signup?sk_intent=playground&provider=CARTESIAMCP)

## Setup

1. ### Install the SDK

   **Node.js**

   ```bash
   npm install @scalekit-sdk/node dotenv
   ```

   **Python**

   ```bash
   pip install scalekit-sdk-python python-dotenv
   ```

2. ### Set your credentials

   Add your Scalekit credentials to your `.env` file. Find values in **[app.scalekit.com](https://app.scalekit.com)** > **Developers** > **API Credentials**.

   ```sh title=".env"
   SCALEKIT_ENVIRONMENT_URL=<your-environment-url>
   SCALEKIT_CLIENT_ID=<your-client-id>
   SCALEKIT_CLIENT_SECRET=<your-client-secret>
   ```

3. ### Create the Cartesia MCP connection

   In **AgentKit > Connections**, create a Cartesia MCP connection. The name you give it is the `connection_name` your code passes. See [Configure connections](/agentkit/connections/).

   **Console steps with screenshots:**

   Register your Cartesia API key with Scalekit so it can authenticate requests to the Cartesia MCP server on behalf of your users. Cartesia MCP uses API key authentication: the key is sent as a Bearer token.

   1. ### Create a Cartesia API key

      - Sign in to the [Cartesia Playground](https://play.cartesia.ai) and open the [API Keys page](https://play.cartesia.ai/keys).
      - Create a new key and give it a descriptive name, such as `Scalekit`.
      - Copy the key.

      > caution: Copy the key now
      >
      > Store the key in a secrets manager. Treat it like a password: anyone with the key can use your Cartesia account and credits.

   2. ### Create a connection in Scalekit

      - In the [Scalekit dashboard](https://app.scalekit.com), go to **AgentKit** → **Connections** → **Create Connection**.
      - Search for **Cartesia MCP** and click **Create**.
      - Note the **Connection name** — use this as `connection_name` in your code (for example, `cartesiamcp`).

   3. ### Add a connected account

      Connected accounts link a specific user identifier in your system to a Cartesia API key. Add one in the dashboard to test. In production, each user adds their own through the authorization link: they enter their credentials on the page it opens.

      **In the dashboard, to test**

      - Open the connection and click the **Connected Accounts** tab → **Add account**.
      - Fill in **Your User's ID** and **API Key**, then click **Save**.

      **From your backend, if your app already has the credentials**

      For example, when users enter them on a settings page in your app:

        **Node.js**

        ```typescript
        import { ConnectorStatus } from '@scalekit-sdk/node'

        const authorizationDetails = {
          details: {
            case: 'staticAuth',
            value: { details: { token: process.env.USER_API_KEY } },
          },
        };

        let { connectedAccount } = await scalekit.actions.upsertConnectedAccount({
          connectionName: 'cartesiamcp',
          identifier: 'user_123',
          authorizationDetails,
        });

        // Make sure the account is ACTIVE before the first tool call.
        if (connectedAccount?.status !== ConnectorStatus.ACTIVE) {
          ({ connectedAccount } = await scalekit.actions.upsertConnectedAccount({
            connectionName: 'cartesiamcp',
            identifier: 'user_123',
            authorizationDetails,
          }));
        }
        ```

        **Python**

        ```python
        authorization_details = {"static_auth": {"token": user_api_key}}

        response = scalekit_client.actions.upsert_connected_account(
            connection_name="cartesiamcp",
            identifier="user_123",
            authorization_details=authorization_details,
        )

        # Make sure the account is ACTIVE before the first tool call.
        if response.connected_account.status != "ACTIVE":
            response = scalekit_client.actions.upsert_connected_account(
                connection_name="cartesiamcp",
                identifier="user_123",
                authorization_details=authorization_details,
            )
        ```

4. ### Authorize a user and make your first call

   **Node.js** (`quickstart.mts`)

   ```typescript
   import { ScalekitClient } from '@scalekit-sdk/node'
   import 'dotenv/config'
   import { createInterface } from 'node:readline/promises'

   const scalekit = new ScalekitClient(
     process.env.SCALEKIT_ENVIRONMENT_URL,
     process.env.SCALEKIT_CLIENT_ID,
     process.env.SCALEKIT_CLIENT_SECRET,
   )
   const actions = scalekit.actions

   const connector = 'cartesiamcp'
   const identifier = 'user_123'

   // Generate an authorization link for the user
   const { link } = await actions.getAuthorizationLink({ connectionName: connector, identifier })
   console.log('Authorize Cartesia MCP:', link)
   const rl = createInterface({ input: process.stdin, output: process.stdout })
   await rl.question('Press Enter after authorizing...')
   rl.close()

   // Make your first call
   const result = await actions.executeTool({
     connector,
     identifier,
     toolName: 'cartesiamcp_list_accents',
     toolInput: {},
   })
   console.log(result)
   ```

   ```bash
   npx tsx quickstart.mts
   ```

   **Python** (`quickstart.py`)

   ```python
   import os
   from scalekit import ScalekitClient
   from dotenv import load_dotenv
   load_dotenv()

   scalekit_client = ScalekitClient(
       env_url=os.getenv("SCALEKIT_ENVIRONMENT_URL"),
       client_id=os.getenv("SCALEKIT_CLIENT_ID"),
       client_secret=os.getenv("SCALEKIT_CLIENT_SECRET"),
   )
   actions = scalekit_client.actions

   connection_name = "cartesiamcp"
   identifier = "user_123"

   # Generate an authorization link for the user
   link_response = actions.get_authorization_link(
       connection_name=connection_name,
       identifier=identifier,
   )
   print("Authorize Cartesia MCP:", link_response.link)
   input("Press Enter after authorizing...")

   # Make your first call
   result = actions.execute_tool(
       tool_input={},
       tool_name="cartesiamcp_list_accents",
       connection_name=connection_name,
       identifier=identifier,
   )
   print(result)
   ```

   ```bash
   python quickstart.py
   ```

   Each user opens the link once and enters their Cartesia MCP server credentials there. If your app already has a user's credentials, add the account from your backend instead, as the console steps above show. See [Authorize a user](/agentkit/tools/authorize/) for the full flow and statuses.

## Tools

Pass the exact name to `execute_tool`, with an input like each tool's example. To ask for a tool that's missing, use the [request form](https://scalekitsupport.portal.usepylon.com/forms/request-a-connector-tool).

**Node.js**

```typescript
const result = await actions.executeTool({
  toolName: 'cartesiamcp_download_file',
  toolInput: {
    file_id: '<file_id>',
  },
  connector: 'cartesiamcp',
  identifier: 'user_123',
})
```

**Python**

```python
result = actions.execute_tool(
    tool_name="cartesiamcp_download_file",
    tool_input={
        "file_id": "<file_id>",
    },
    connection_name="cartesiamcp",
    identifier="user_123",
)
```

`result.data` is the app's response as JSON, and `result.execution_id` (`executionId` in Node.js) is the ID of the call.

### `cartesiamcp_download_file`

Download file · Read-only

        Fetch a Cartesia cloud-stored file by ID. Returns a 24-hour `download_url` for
        hosted and remote clients. Also writes a copy to `OUTPUT_DIRECTORY` on the MCP
        server host (for local `uvx` or server-side `speech_to_text` in the same session).
        On hosted MCP those local copies are under a temp directory and purged after a TTL.

        Use the `file_id` from a prior `text_to_speech` call (`save=true`, the default)
        or from playground TTS history.

        Parameters
        ----------
        file_id : str
            Cloud file ID (e.g. from `text_to_speech`).

        format : typing.Optional[DownloadFormat]
            Pass `playback` to wrap raw PCM as WAV in the download link and local file.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `file_id` (`string`, required): Cloud file ID (e.g. from `text_to_speech`).
- `format` (`string`, optional): Pass `playback` to wrap raw PCM as WAV in the download link and local file. One of: `playback`.

Example input: `{"file_id":"<file_id>"}`

### `cartesiamcp_get_pronunciation_dict`

Get pronunciation dictionary · Read-only

        Parameters
        ----------
        dict_id : str
            Pronunciation dictionary ID.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `dict_id` (`string`, required): Pronunciation dictionary ID.

Example input: `{"dict_id":"<dict_id>"}`

### `cartesiamcp_get_voice`

Get voice · Read-only

        Parameters
        ----------
        voice_id : str
            The ID of the voice to get.

        request_options : typing.Optional[RequestOptions]
            Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `voice_id` (`string`, required): The ID of the voice to get.
- `request_options` (`object`, optional): Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Example input: `{"voice_id":"<voice_id>"}`

### `cartesiamcp_list_accents`

List accents · Read-only

        List the official accent catalog from GET /accents.

        Each entry includes the canonical accent `id` plus display metadata
        (`name`, `language`, `locale`, `is_locale_default`, `is_localizable`).
        Pass `id` to `clone_voice`, `add_voice_accents`, `delete_voice_accent`,
        and `localize_voice`. Display names are rejected on this API version.

        Parameters
        ----------
        request_options : typing.Optional[RequestOptions]
            Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `request_options` (`object`, optional): Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Example input: `{}`

### `cartesiamcp_list_agent_models`

List agent models · Read-only

        List LLMs available for Managed Agents from GET /v1/agents/models.

        Use these `id` values for an agent's `model.id` (and explicit fallback
        model ids). Do not invent model ids. Each row includes provider,
        display name, latency, and token pricing. Returns the full catalog
        (pagination is followed server-side).

        Parameters
        ----------
        request_options : typing.Optional[RequestOptions]
            Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `request_options` (`object`, optional): Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Example input: `{}`

### `cartesiamcp_list_models`

List models · Read-only

        List callable Sonic (TTS) and Ink (STT) model ids from GET /models.

        Each row has `id` and `capability` (`tts` or `stt`). TTS rows include
        `supported_languages`, the sorted synthesis language codes that model accepts.
        STT rows include `modes` (`stream` and/or `batch`). Pass `id` to `text_to_speech` as
        `model_id` or to `speech_to_text` as `model`. Omit model on those tools
        when you want the API default. Preview ids (`sonic-preview`, `ink-preview`)
        are included when callable.

        Parameters
        ----------
        capability : typing.Optional[Literal["tts", "stt"]]
            Filter to TTS or STT. Omit to return both.

        request_options : typing.Optional[RequestOptions]
            Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `capability` (`string`, optional): Filter to TTS or STT. Omit to return both. One of: `tts`, `stt`.
- `request_options` (`object`, optional): Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Example input: `{}`

### `cartesiamcp_list_pronunciation_dicts`

List pronunciation dictionaries · Read-only

        List pronunciation dictionaries for the authenticated user.

        Parameters
        ----------
        limit : typing.Optional[int]
            Number of dictionaries per page (1–100).

        starting_after : typing.Optional[str]
            Cursor: dictionary ID to start after.

        ending_before : typing.Optional[str]
            Cursor: dictionary ID to end before.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `ending_before` (`string`, optional): Cursor: dictionary ID to end before.
- `limit` (`integer`, optional): Number of dictionaries per page (1–100). Default: `10`.
- `starting_after` (`string`, optional): Cursor: dictionary ID to start after.

Example input: `{}`

### `cartesiamcp_list_voices`

List voices · Read-only

        Parameters
        ----------
        limit : typing.Optional[int]
            The number of Voices to return per page, ranging between 1 and 100.

        starting_after : typing.Optional[str]
            A cursor to use in pagination. `starting_after` is a Voice ID that defines your
            place in the list. For example, if you make a /voices request and receive 100
            objects, ending with `voice_abc123`, your subsequent call can include
            `starting_after=voice_abc123` to fetch the next page of the list.

        ending_before : typing.Optional[str]
            A cursor to use in pagination. `ending_before` is a Voice ID that defines your
            place in the list. For example, if you make a /voices request and receive 100
            objects, starting with `voice_abc123`, your subsequent call can include
            `ending_before=voice_abc123` to fetch the previous page of the list.

        is_owner : typing.Optional[bool]
            Whether to only return voices owned by the current user.

        is_starred : typing.Optional[bool]
            Whether to only return starred voices.

        gender : typing.Optional[GenderPresentation]
            The gender presentation of the voices to return.

        language : typing.Optional[str]
            Filter voices by language or locale, such as `en`, `it`, or `en_GB`.
            A locale returns accents for that region; a language alone returns all accents
            for that language. Both `-` and `_` separators are accepted.

        q : typing.Optional[str]
            Search voices by name, description, or voice ID.

        expand : typing.Optional[typing.Sequence[str]]
            Additional fields to include in the response, such as `preview_file_url`.

        request_options : typing.Optional[RequestOptions]
            Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `ending_before` (`string`, optional): A cursor to use in pagination. `ending_before` is a Voice ID that defines your place in the list. For example, if you make a /voices request and receive 100 objects, starting with `voice_abc123`, your subsequent call can include `ending_before=voice_abc123` to fetch the previous page of the list.
- `expand` (`array`, optional): Additional fields to include in the response, such as `preview_file_url`.
- `gender` (`string`, optional): The gender presentation of the voices to return. One of: `masculine`, `feminine`, `gender_neutral`.
- `is_owner` (`boolean`, optional): Whether to only return voices owned by the current user.
- `is_starred` (`boolean`, optional): Whether to only return starred voices.
- `language` (`string`, optional): Filter voices by language or locale, such as `en`, `it`, or `en_GB`. A locale returns accents for that region; a language alone returns all accents for that language. Both `-` and `_` separators are accepted.
- `limit` (`integer`, optional): The number of Voices to return per page, ranging between 1 and 100. Default: `10`.
- `q` (`string`, optional): Search voices by name, description, or voice ID.
- `request_options` (`object`, optional): Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).
- `starting_after` (`string`, optional): A cursor to use in pagination. `starting_after` is a Voice ID that defines your place in the list. For example, if you make a /voices request and receive 100 objects, ending with `voice_abc123`, your subsequent call can include `starting_after=voice_abc123` to fetch the next page of the list.

Example input: `{}`

### `cartesiamcp_speech_to_text`

Transcribe audio · Read-only

        Transcribe a pre-recorded audio file to text.

        **Default (`mode="batch"`):** upload the file via batch STT (`POST /stt`).
        Best for typical MCP file-on-disk tasks (mp3, flac, wav, and other containers).
        Default model: `ink-whisper`.

        **Streaming (`mode="stream"`):** send mono PCM over STT WebSocket (`/stt/websocket`).
        Use for mono PCM WAV (for example TTS output) or raw PCM with `encoding` and
        `sample_rate`. Default model: `ink-2`.

        **Pricing:** See [STT pricing](https://docs.cartesia.ai/pricing#speech-to-text).

        Pass `file_id` or `file_path`, not both. On hosted MCP, pass the `file_id`
        from `text_to_speech` or `download_file`. Do not pass a path from the agent
        machine — `file_path` is an absolute path on this MCP server and must exist.
        A missing path on hosted MCP returns a validation error telling you to use
        `file_id`.

        Parameters
        ----------
        file_path : typing.Optional[str]
            Absolute path on this MCP server. Must exist; hosted MCP rejects client laptop paths.

        mode : str
            `batch` (default) or `stream`.

        model : typing.Optional[str]
            STT model ID. Defaults to `ink-whisper` for batch and `ink-2` for
            stream. Call `list_models` with `capability="stt"` for valid ids.

        language : typing.Optional[str]
            ISO-639-1 language code (defaults to `en` for stream; batch uses API default).

        encoding : typing.Optional[SttEncoding]
            Required for raw PCM without a container header (both modes).

        sample_rate : typing.Optional[int]
            Sample rate in Hz when `encoding` is set.

        timestamp_granularities : typing.Optional[typing.Sequence[TimestampGranularity]]
            Pass `["word"]` for word-level timestamps when supported.

        file_id : typing.Optional[str]
            Cartesia cloud file id. The server downloads it. Use this from hosted clients.

        request_options : typing.Optional[RequestOptions]
            Request-specific configuration for batch mode (timeout, headers, query params, extra body fields, max_retries). Ignored in stream mode.

Read only: yes. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `encoding` (`string`, optional): Required for raw PCM without a container header (both modes). One of: `pcm_s16le`, `pcm_s32le`, `pcm_f16le`, `pcm_f32le`, `pcm_mulaw`, `pcm_alaw`.
- `file_id` (`string`, optional): Cartesia cloud file id. The server downloads it. Use this from hosted clients.
- `file_path` (`string`, optional): Absolute path on this MCP server. Must exist; hosted MCP rejects client laptop paths.
- `language` (`string`, optional): ISO-639-1 language code (defaults to `en` for stream; batch uses API default).
- `mode` (`string`, optional): `batch` (default) or `stream`. One of: `batch`, `stream`. Default: `batch`.
- `model` (`string`, optional): STT model ID. Defaults to `ink-whisper` for batch and `ink-2` for stream. Call `list_models` with `capability="stt"` for valid ids.
- `request_options` (`object`, optional): Request-specific configuration for batch mode (timeout, headers, query params, extra body fields, max_retries). Ignored in stream mode.
- `sample_rate` (`integer`, optional): Sample rate in Hz when `encoding` is set.
- `timestamp_granularities` (`array`, optional): Pass `["word"]` for word-level timestamps when supported.

Example input: `{}`

### `cartesiamcp_add_voice_accents`

Add voice accents · Write

        Add accents to an instant voice clone you own so a single voice_id can speak
        multiple accents natively. A voice can support up to 10 accents in total,
        including native.

        Parameters
        ----------
        voice_id : str
            The ID of the voice.

        accents : list[VoiceAccent]
            Catalog accent ids from GET /accents (for example `british` or `parisian`).
            Display names are rejected on this API version.

        request_options : typing.Optional[RequestOptions]
            Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `accents` (`array`, required): Catalog accent ids from GET /accents (for example `british` or `parisian`). Display names are rejected on this API version.
- `voice_id` (`string`, required): The ID of the voice.
- `request_options` (`object`, optional): Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Example input: `{"accents":[],"voice_id":"<voice_id>"}`

### `cartesiamcp_clone_voice`

Clone voice · Write

        Clone a voice from an audio clip. This endpoint has two modes, stability and similarity.

        Similarity mode clones are more similar to the source clip, but may reproduce background noise. For these, use an audio clip about 5 seconds long.

        Stability mode clones are more stable, but may not sound as similar to the source clip. For these, use an audio clip 10-20 seconds long.

        Pass `file_id` or `file_path`, not both. On hosted MCP, pass the `file_id`
        from `text_to_speech` or `download_file`. `file_path` is an absolute path on
        this MCP server (local `uvx`, or a path returned in this same session). A path
        on the agent machine is not visible to hosted MCP and returns a validation
        error — use `file_id` instead.

        Parameters
        ----------
        name : str
            The name of the voice.

        language : SupportedLanguage
            The language of the voice.

        mode : CloneMode
            Tradeoff between similarity and stability. Similarity clones sound more like the source clip, but may reproduce background noise. Stability clones always sound like a studio recording, but may not sound as similar to the source clip.

        file_path : typing.Optional[str]
            Absolute path on this MCP server. Must exist; hosted MCP rejects client laptop paths.

        file_id : typing.Optional[str]
            Cartesia cloud file id. The server downloads it.

        description : typing.Optional[str]
            A description for the voice.

        accent : typing.Optional[VoiceAccent]
            Catalog accent id from GET /accents (for example `southern-us` or `parisian`).
            Display names are rejected on this API version. Omit to leave unset.

        request_options : typing.Optional[RequestOptions]
            Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `language` (`string`, required): The language of the voice.
- `mode` (`string`, required): Tradeoff between similarity and stability. Similarity clones sound more like the source clip, but may reproduce background noise. Stability clones always sound like a studio recording, but may not sound as similar to the source clip.
- `name` (`string`, required): The name of the voice.
- `accent` (`string`, optional): Catalog accent id from GET /accents (for example `southern-us` or `parisian`). Display names are rejected on this API version. Omit to leave unset. One of: `abruzzo-italian`, `african-american`, `african-french`, `arabic`, `arabic-english`, `australian`, `bagheli`, `belgian-french`, `brazilian-portuguese`, `british`, `budapest`, `bulgarian`, `california`, `camba`, `campania`, `canadian-english`, `canadian-french`, `castilian`, `central-tamil`, `central-thai`, `central-vietnamese`, `chilean`, `colombian`, `czech`, `danish`, `european-portuguese`, `finnish`, `general-american`, `high-german`, `hindi`, `indian-english`, `indian-urdu`, `irish`, `israeli`, `istanbul`, `italian`, `jakarta`, `japanese`, `jessore`, `khaleeji`, `konkani`, `korean`, `kyiv`, `malay`, `mandarin`, `manila`, `mexican`, `middle-eastern-arabic`, `midwestern-american`, `modern-standard-arabic`, `moldovan`, `new-york`, `new-zealand`, `north-kerala`, `odia`, `oslo`, `parisian`, `parsi`, `peruvian`, `polish`, `powadhi`, `randstad`, `romanian`, `russian`, `singaporean`, `slovak`, `south-african`, `southern-us`, `southern-karnataka`, `southern-vietnamese`, `standard-japanese`, `stockholm`, `swiss-standard`, `taiwanese-mandarin`, `tbilisi`, `telangana`, `urdu`, `thessaloniki`, `zagreb`.
- `description` (`string`, optional): A description for the voice.
- `file_id` (`string`, optional): Cartesia cloud file id. The server downloads it.
- `file_path` (`string`, optional): Absolute path on this MCP server. Must exist; hosted MCP rejects client laptop paths.
- `request_options` (`object`, optional): Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Example input: `{"language":"<language>","mode":"<mode>","name":"<name>"}`

### `cartesiamcp_create_pronunciation_dict`

Create pronunciation dictionary · Write

        Create a pronunciation dictionary.

        Parameters
        ----------
        name : str

        items : typing.Optional[typing.Sequence[PronunciationDictItemParams]]
            Mappings of `text` to `pronunciation` (IPA or sounds-like).
            Optional `case_sensitive` (default false): when false, match every
            capitalization of `text` on Sonic 3.6.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `name` (`string`, required): Name of the pronunciation dictionary.
- `items` (`array`, optional): Mappings of `text` to `pronunciation` (IPA or sounds-like). Optional `case_sensitive` (default false): when false, match every capitalization of `text` on Sonic 3.6.

Example input: `{"name":"<name>"}`

### `cartesiamcp_localize_voice`

Localize voice · Write

        Create a new voice from an existing voice localized to a new language and dialect.

        Parameters
        ----------
        voice_id : str
            The ID of the voice to localize.

        name : str
            The name of the new localized voice.

        description : str
            The description of the new localized voice.

        accent : VoiceAccent
            Catalog accent id from GET /accents (for example `mexican` or `parisian`).
            Display names are rejected on this API version.

        language : LocalizeTargetLanguage
            Target language to localize the voice to. Distinct from TTS SupportedLanguage.

        original_speaker_gender : Gender

        dialect : typing.Optional[LocalizeDialect]
            Dialect allowlist for English, Spanish, Portuguese, and French.

        request_options : typing.Optional[RequestOptions]
            Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `accent` (`string`, required): Catalog accent id from GET /accents (for example `mexican` or `parisian`). Display names are rejected on this API version. One of: `abruzzo-italian`, `african-american`, `african-french`, `arabic`, `arabic-english`, `australian`, `bagheli`, `belgian-french`, `brazilian-portuguese`, `british`, `budapest`, `bulgarian`, `california`, `camba`, `campania`, `canadian-english`, `canadian-french`, `castilian`, `central-tamil`, `central-thai`, `central-vietnamese`, `chilean`, `colombian`, `czech`, `danish`, `european-portuguese`, `finnish`, `general-american`, `high-german`, `hindi`, `indian-english`, `indian-urdu`, `irish`, `israeli`, `istanbul`, `italian`, `jakarta`, `japanese`, `jessore`, `khaleeji`, `konkani`, `korean`, `kyiv`, `malay`, `mandarin`, `manila`, `mexican`, `middle-eastern-arabic`, `midwestern-american`, `modern-standard-arabic`, `moldovan`, `new-york`, `new-zealand`, `north-kerala`, `odia`, `oslo`, `parisian`, `parsi`, `peruvian`, `polish`, `powadhi`, `randstad`, `romanian`, `russian`, `singaporean`, `slovak`, `south-african`, `southern-us`, `southern-karnataka`, `southern-vietnamese`, `standard-japanese`, `stockholm`, `swiss-standard`, `taiwanese-mandarin`, `tbilisi`, `telangana`, `urdu`, `thessaloniki`, `zagreb`.
- `description` (`string`, required): The description of the new localized voice.
- `language` (`string`, required): Target language to localize the voice to. Distinct from TTS SupportedLanguage. One of: `en`, `de`, `es`, `fr`, `ja`, `pt`, `zh`, `hi`, `it`, `ko`, `nl`, `pl`, `ru`, `sv`, `tr`, `ar`, `he`, `ta`, `te`, `th`.
- `name` (`string`, required): The name of the new localized voice.
- `original_speaker_gender` (`string`, required): Original speaker gender. One of: `male`, `female`.
- `voice_id` (`string`, required): The ID of the voice to localize.
- `dialect` (`string`, optional): Dialect allowlist for English, Spanish, Portuguese, and French. One of: `au`, `in`, `so`, `uk`, `us`, `mx`, `pe`, `br`, `eu`, `ca`.
- `request_options` (`object`, optional): Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Example input: `{"accent":"abruzzo-italian","description":"<description>","language":"en","name":"<name>","original_speaker_gender":"male","voice_id":"<voice_id>"}`

### `cartesiamcp_text_to_speech`

Convert text to speech · Write

        Generate speech audio from text. By default (`save=true`) the audio is persisted
        in Cartesia cloud storage and the response includes `file_id` and `download_url`
        (24-hour public link). Pass that `file_id` to `speech_to_text` or `clone_voice`
        — including from a hosted client whose disk is not this server.
        `download_url` is a browser link. On the local server, `file_path` is a copy under
        `OUTPUT_DIRECTORY` (or `output_path`). On hosted MCP with `save=true` (the default),
        `file_path` is omitted — use `file_id` / `download_url` instead.

        Parameters
        ----------
        transcript : str

        voice_id : str
            Cartesia voice ID (e.g. from `list_voices`).

        output_format : OutputFormatParams

        model_id : str
            The ID of the model to use for the generation. Call `list_models`
            with `capability="tts"` for valid ids.

        language : typing.Optional[SupportedLanguage]

        duration : typing.Optional[float]
            The maximum duration of the audio in seconds. You do not usually need to specify this.
            If the duration is not appropriate for the length of the transcript, the output audio may be truncated.

        speed : typing.Optional[float]
            Speech speed multiplier (0.6–1.5). Applies to Sonic 3+ via `generation_config`.

        volume : typing.Optional[float]
            Volume multiplier (0.5–2.0). Applies to Sonic 3+ via `generation_config`.

        emotion : typing.Optional[str]
            Emotional guidance for the generation (e.g. `neutral`, `excited`, `sad`). Applies to Sonic 3+.

        pronunciation_dict_id : typing.Optional[str]
            Pronunciation dictionary ID to apply for this generation only.

        output_path : typing.Optional[str]
            Local file path to write the audio (for example `src/assets/audio/click.wav`).
            Relative paths resolve against the MCP server working directory. Only supported
            with the local server (stdio / uvx); rejected on hosted MCP.

        save : bool
            When true (default), persist the generation to cloud storage and return `file_id`
            plus `download_url`. Pass `download_file` with that `file_id` to refresh the link
            or write another local copy.

        request_options : typing.Optional[RequestOptions]
            Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `output_format` (`string`, required): Output audio format. One of a raw PCM, WAV, or MP3 container, each with its own required settings (encoding/bit rate and sample rate).
- `transcript` (`string`, required): The text to convert to speech.
- `voice_id` (`string`, required): Cartesia voice ID (e.g. from `list_voices`).
- `duration` (`number`, optional): The maximum duration of the audio in seconds. You do not usually need to specify this. If the duration is not appropriate for the length of the transcript, the output audio may be truncated.
- `emotion` (`string`, optional): Emotional guidance for the generation (e.g. `neutral`, `excited`, `sad`). Applies to Sonic 3+.
- `language` (`string`, optional): Language code.
- `model_id` (`string`, optional): The ID of the model to use for the generation. Call `list_models` with `capability="tts"` for valid ids. Default: `sonic-3.5`.
- `output_path` (`string`, optional): Local file path to write the audio (for example `src/assets/audio/click.wav`). Relative paths resolve against the MCP server working directory. Only supported with the local server (stdio / uvx); rejected on hosted MCP.
- `pronunciation_dict_id` (`string`, optional): Pronunciation dictionary ID to apply for this generation only.
- `request_options` (`object`, optional): Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).
- `save` (`boolean`, optional): When true (default), persist the generation to cloud storage and return `file_id` plus `download_url`. Pass `download_file` with that `file_id` to refresh the link or write another local copy. Default: `true`.
- `speed` (`number`, optional): Speech speed multiplier (0.6–1.5). Applies to Sonic 3+ via `generation_config`.
- `volume` (`number`, optional): Volume multiplier (0.5–2.0). Applies to Sonic 3+ via `generation_config`.

Example input: `{"output_format":"<output_format>","transcript":"<transcript>","voice_id":"<voice_id>"}`

### `cartesiamcp_update_pronunciation_dict`

Update pronunciation dictionary · Write

        Update a pronunciation dictionary.

        Parameters
        ----------
        dict_id : str

        name : typing.Optional[str]

        items : typing.Optional[typing.Sequence[PronunciationDictItemParams]]
            Optional `case_sensitive` (default false) on each item.

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `dict_id` (`string`, required): ID of the pronunciation dictionary to update.
- `items` (`array`, optional): Pronunciation items (text to pronunciation mappings). Optional `case_sensitive` (default false) on each item.
- `name` (`string`, optional): New name for the pronunciation dictionary.

Example input: `{"dict_id":"<dict_id>"}`

### `cartesiamcp_update_voice`

Update voice · Write

        Parameters
        ----------
        id : VoiceId

        name : str
            The name of the voice.

        description : str
            The description of the voice.

        request_options : typing.Optional[RequestOptions]
            Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Read only: no. Destructive: no. Idempotent (safe to retry): no.

Inputs:

- `description` (`string`, required): The description of the voice.
- `name` (`string`, required): The name of the voice.
- `voice_id` (`string`, required): The ID of the voice to update.
- `request_options` (`object`, optional): Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Example input: `{"description":"<description>","name":"<name>","voice_id":"<voice_id>"}`

### `cartesiamcp_delete_pronunciation_dict`

Delete pronunciation dictionary · Destructive

        Parameters
        ----------
        dict_id : str
            Pronunciation dictionary ID to delete.

Read only: no. Destructive: yes. Idempotent (safe to retry): no.

Inputs:

- `dict_id` (`string`, required): Pronunciation dictionary ID to delete.

Example input: `{"dict_id":"<dict_id>"}`

### `cartesiamcp_delete_voice`

Delete voice · Destructive

        Parameters
        ----------
        voice_id : str
            The ID of the voice to delete.

        request_options : typing.Optional[RequestOptions]
            Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Read only: no. Destructive: yes. Idempotent (safe to retry): no.

Inputs:

- `voice_id` (`string`, required): The ID of the voice to delete.
- `request_options` (`object`, optional): Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Example input: `{"voice_id":"<voice_id>"}`

### `cartesiamcp_delete_voice_accent`

Delete voice accent · Destructive

        Remove an accent an instant voice clone supports.

        Parameters
        ----------
        voice_id : str
            The ID of the voice.

        accent : VoiceAccent
            Catalog accent id from GET /accents (for example `british` or `parisian`).
            Display names are rejected on this API version.

        request_options : typing.Optional[RequestOptions]
            Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Read only: no. Destructive: yes. Idempotent (safe to retry): no.

Inputs:

- `accent` (`string`, required): Catalog accent id from GET /accents (for example `british` or `parisian`). Display names are rejected on this API version. One of: `abruzzo-italian`, `african-american`, `african-french`, `arabic`, `arabic-english`, `australian`, `bagheli`, `belgian-french`, `brazilian-portuguese`, `british`, `budapest`, `bulgarian`, `california`, `camba`, `campania`, `canadian-english`, `canadian-french`, `castilian`, `central-tamil`, `central-thai`, `central-vietnamese`, `chilean`, `colombian`, `czech`, `danish`, `european-portuguese`, `finnish`, `general-american`, `high-german`, `hindi`, `indian-english`, `indian-urdu`, `irish`, `israeli`, `istanbul`, `italian`, `jakarta`, `japanese`, `jessore`, `khaleeji`, `konkani`, `korean`, `kyiv`, `malay`, `mandarin`, `manila`, `mexican`, `middle-eastern-arabic`, `midwestern-american`, `modern-standard-arabic`, `moldovan`, `new-york`, `new-zealand`, `north-kerala`, `odia`, `oslo`, `parisian`, `parsi`, `peruvian`, `polish`, `powadhi`, `randstad`, `romanian`, `russian`, `singaporean`, `slovak`, `south-african`, `southern-us`, `southern-karnataka`, `southern-vietnamese`, `standard-japanese`, `stockholm`, `swiss-standard`, `taiwanese-mandarin`, `tbilisi`, `telangana`, `urdu`, `thessaloniki`, `zagreb`.
- `voice_id` (`string`, required): The ID of the voice.
- `request_options` (`object`, optional): Request-specific configuration (timeout, headers, query params, extra body fields, max_retries).

Example input: `{"accent":"abruzzo-italian","voice_id":"<voice_id>"}`

## Related

Other AI connectors ([all 216](/agentkit/connectors/?category=ai)):

- [Chorus](/agentkit/connectors/chorus/): Scalekit connector, Username and password, 9 tools
- [Gong](/agentkit/connectors/gong/): Scalekit connector, OAuth, 69 tools
- [Jiminny](/agentkit/connectors/jiminny/): Scalekit connector, Access token, 27 tools


---

## More Scalekit documentation

| Resource | What it contains | When to use it |
|----------|-----------------|----------------|
| [/llms.txt](/llms.txt) | Structured index with routing hints per product area | Start here — find which documentation set covers your topic before loading full content |
| [/llms-full.txt](/llms-full.txt) | Complete documentation for all Scalekit products in one file | Use when you need exhaustive context across multiple products or when the topic spans several areas |
| [sitemap-0.xml](https://docs.scalekit.com/sitemap-0.xml) | Full URL list of every documentation page | Use to discover specific page URLs you can fetch for targeted, page-level answers |
