Can Lyro delete any voice on the LMNT account?

Only voices you own. Delete Voice Info works where owner is 'me' and cannot remove system voices, and the deletion is permanent and cancels any pending operations on that voice - which makes it the one action worth gating behind human approval.

Is a newly created custom voice usable straight away?

Not always. Create Voice returns a voice ID immediately, but the voice can come back in a training state before it turns ready, so Lyro checks Get Voice Info for active status and supported languages before sending a synthesis request against it.

What comes back from a synthesis request?

Base64-encoded audio at data.response_data.audio, which has to be decoded before it is saved or handed to another tool. Synthesize Speech is the lower-latency call; Generate Speech With Metadata also returns the synthesis seed and word-level timings.

Do all LMNT voices work on every subscription plan?

Get Voice Info returns plan availability alongside active status and supported languages for each voice, so Lyro can confirm a voice is usable on your subscription before calling it, and Get Account reports the plan and current usage behind that check.

Lmnt MCP integration

Synthesizes speech in preset or custom voices, returns word-level timings for captions, and keeps the voice library and usage in check.

Connect Lmnt See all integrations

8actions available

Three actions you can hand over today

Every action runs live through MCP. Nothing to build, nothing to maintain.

Synthesize speech
Lyro speaks support responses in natural-sounding voices. Customers hear personalized audio messages instead of only reading text.
Create voice
Lyro builds custom voices trained on sample audio. Your team creates branded voice personas for consistent, recognizable customer interactions.
Get voices list
Lyro retrieves available voices for different response types. Your team selects voices matching tone and personality for each customer scenario.

See all 8 actions

How businesses use Lmnt + Lyro

Each card is one request a support team gets, and the Lmnt actions Lyro runs to close it.

Speak a reply, with timings when captions need them
Lyro synthesizes up to 5,000 characters into audio in a chosen voice, or switches to the metadata variant when word-level timestamps are needed to sync subtitles.
Synthesize SpeechGenerate Speech With MetadataGet Voice Info
Check the plan before a long synthesis run
Lyro reads the account's subscription plan and current usage statistics, so a large batch is measured against what the plan allows rather than discovered at the invoice.
Get AccountGet Voices ListGet Voice Info
Train a custom voice from uploaded samples
Lyro creates a voice from audio samples and reports when it moves from training to ready, so a brand voice becomes available for synthesis without a console round trip.
Create VoiceGet Voice InfoGet Voices List
Keep the voice library down to what is used
Lyro renames a voice, corrects its description, stars the ones in rotation, and deletes a custom voice nobody uses, separating system presets from what your team made.
Update Voice InfoDelete Voice InfoGet Voices List

How it works

Get started in 3 steps

Connect once, then just ask. There is no workflow builder to learn and nothing to maintain — Lyro reads the Lmnt actions it has and picks the ones a request needs.

01
Connect Lmnt
Authorize the Lmnt account your team already uses — one consent screen, no API keys, no mapping tables. Lyro can only do what you granted that account, and you can disconnect it at any time.
02
Tell your agent what you need
Describe the job the way you would hand it to a teammate. Lyro maps it to the Lmnt actions that close it and chains as many as the request needs.
03
Watch it work
The agent runs the actions inside the conversation the customer is already in, so nobody copies data between tabs and your team can take over at any point.
Get started free

Lmnt · Lyro

Everything else about Lmnt

Setup, permissions, and the limits of what Lyro can do inside Lmnt.

Only voices you own. Delete Voice Info works where owner is 'me' and cannot remove system voices, and the deletion is permanent and cancels any pending operations on that voice - which makes it the one action worth gating behind human approval.

Every action available in Lmnt

All 8 actions your agent can call on Lmnt, straight from the live MCP connection.

Create voice
Creates a custom voice in LMNT by training on uploaded audio samples.
Delete voice info
Deletes a voice from your LMNT account.
Generate speech with metadata
Generates speech from text and returns JSON with base64-encoded audio and optional word-level timing metadata.
Get account
Retrieves account information including subscription plan details and current usage statistics.
Get voice info
Gets metadata for a specific LMNT voice, including active status, supported languages, and plan availability.
Get voices list
Retrieves a list of available voices from LMNT.
Synthesize speech
Synthesizes speech from text using LMNT's AI voices.
Update voice info
Updates information about a specific voice in LMNT.