VoiceLabs

Clone a voice from a recording

clone_voice_profile

Create a cloned voice on the user's VoiceLabs account from a recording the caller provides: a name, the language, the exact text spoken in the recording, and the audio either inline as base64 (audioBase64) or as an https:// URL (audioUrl). The recording must be 2–30 seconds of clear speech. Returns the new profile's id, name, voice type and engine; pass the id to speak. Not idempotent: a second call with the same name is refused, so check list_voice_profiles first. Only clone a voice whose owner has agreed to it.

Input Schema

Name	Required	Description
`name`	Yes	The name for the new cloned voice. Must not already exist on the account — call list_voice_profiles first if unsure; a duplicate is refused with INVALID_REQUEST.
`audioUrl`	No	An https:// URL the recording can be downloaded from (public host, no redirects, up to 10 MiB). Give exactly one of audioBase64 or audioUrl.
`language`	No	ISO language code of the recording and the voice. Defaults to "en".
`audioBase64`	No	The recording inline as base64 (WAV, or any of mp3/m4a/ogg/flac/aac/webm/opus; up to 10 MiB decoded). 2–30 seconds of clear speech. Give exactly one of audioBase64 or audioUrl.
`referenceText`	Yes	The exact words spoken in the recording (up to 1000 characters). The engine aligns the clone against this transcript, so it must match what is said.

Output Schema

Name	Required	Description	Default
`id`	Yes
`name`	Yes
`engine`	Yes
`language`	Yes
`sampleId`	Yes
`voiceType`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

Changed2 schema fields changedOct 26, 2026
- removedOutput schema / properties / engine / anyOf
```
Removed value: -[
-  {
-    "type": "string"
-  },
-  {
-    "type": "null"
-  }
-]
```
- addedOutput schema / properties / engine / type
```
Added value: +[
+  "string",
+  "null"
+]
```
First observedSep 28, 2026

TDQS

A4.4/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare idempotentHint=false and destructiveHint=false, but the description adds why the non-idempotency matters in practice — a same-name second call is refused, not merged — plus the consent requirement and the duration constraint. It also names the returned fields, though that is partly redundant with the output schema.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with the action, and every later sentence earns its place (duration limit, return shape, idempotency warning, consent). The opening sentence does re-enumerate the parameters already covered at 100% in the schema, which is mild duplication.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 5-parameter, non-idempotent, open-world mutation, the description covers what is created, what input is required, what comes back, and what could go wrong. An output schema exists, so the brief return mention is sufficient rather than necessary.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so all five parameters are already documented in depth (including the exactly-one-of constraint on audioBase64/audioUrl). The description restates the same parameter list and the either/or audio rule without adding format or constraint detail beyond the schema. Baseline 3.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Create a cloned voice on the user's VoiceLabs account from a recording') and immediately disambiguates from siblings by naming the follow-up ('pass the id to speak') and the prerequisite ('check list_voice_profiles first'). An agent can distinguish this from ensure_voice_profile or speak without opening a schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives explicit preconditions (2–30 seconds of clear speech, owner consent), an explicit pre-check (list_voice_profiles to avoid duplicate-name refusal), and the downstream handoff (speak). Both when-to-use and the failure-avoidance path are spelled out.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All VoiceLabs tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?