Skip to main content
Glama
Sign In

switch

by ai.switchapp
OverviewDiscussionsAdmin

Generate Speech

generate_speech

Speak written words out loud in an ElevenLabs voice and save the audio to your library. Give it the words and the voice (a voice name, or a voice id from list_voices). Every take speaks on Eleven v4 or Eleven v4 Turbo. Choose with quality: "fast" (the default, v4 Turbo, quick and cheap) or "best" (v4, the richest, most emotional read), or name one with model: v4 or turbo. v4 holds 10,000 letters a take and turbo has no stated limit. Both take stability and similarity (0 to 1) only. The price is checked before generation and the result includes its billing receipt. The answer carries the time in seconds that every single word starts and ends, which is what you use to cut on the beat, place captions or time a picture to a word. For a sound effect use generate_sound_effect; for two or more people talking in one file use generate_dialogue.

Input Schema

NameRequiredDescriptionDefault
textYesThe words to speak, exactly as they should be said. How long one take can be is set by the model you pick. Write numbers and dates the way you want them read out.
modelNoOptional. v4 (richest) or turbo (quick, the default). Leave it out and quality decides.
speedNoNot used: Eleven v4 has no speed dial, so this is ignored.
styleNoNot used: Eleven v4 has no style dial, so this is ignored.
voiceYesWhich voice speaks: a voice name such as "Rachel", or a voice id. Call list_voices to see the voices on this account.
qualityNoOptional. "best" picks the richest model, "fast" the quick one. Default fast.
languageNoOptional two-letter language code, for example "es" or "fr", when the words are not in English.
stabilityNoOptional, 0 to 1. Lower is more expressive and varied, higher is steadier and more even.
request_idNoOnly reuse a request_id explicitly returned by an enabled durable recovery workflow; an attempt_id or saved-file recovery receipt does not authorize this replay.
similarityNoOptional, 0 to 1. How closely the read sticks to the original voice.
timestampsNoOptional, default true. Word-by-word times come back with the audio. Set false if you do not need them.
speaker_boostNoNot used: Eleven v4 has no speaker boost, so this is ignored.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • addedInput schema / properties / request_id
      Added value: +{
+  "description": "Only reuse a request_id explicitly returned by an enabled durable recovery workflow; an attempt_id or saved-file recovery receipt does not authorize this replay.",
+  "type": "string"
+}
  2. Changed4 schema fields changed
    • changedInput schema / properties / model / description
      Previous value: -"Optional. v4 (richest), turbo (quick, the default), v3, conversational, multilingual or flash. Leave it out and quality decides."New value: +"Optional. v4 (richest) or turbo (quick, the default). Leave it out and quality decides."
    • changedInput schema / properties / speaker_boost / description
      Previous value: -"Optional. Holds the voice closer to the speaker it was made from. Every model except v4 and v4 turbo."New value: +"Not used: Eleven v4 has no speaker boost, so this is ignored."
    • changedInput schema / properties / speed / description
      Previous value: -"Optional, 0.7 to 1.2. Under 1 is slower, over 1 is faster. Every model except v4 and v4 turbo."New value: +"Not used: Eleven v4 has no speed dial, so this is ignored."
    • changedInput schema / properties / style / description
      Previous value: -"Optional, 0 to 1. Extra performance. Every model except v4 and v4 turbo."New value: +"Not used: Eleven v4 has no style dial, so this is ignored."
  3. Changed5 schema fields changed
    • changedInput schema / properties / model / description
      Previous value: -"Optional. v4 (richest), turbo (quick, the default), v3, multilingual or flash. Leave it out and quality decides."New value: +"Optional. v4 (richest), turbo (quick, the default), v3, conversational, multilingual or flash. Leave it out and quality decides."
    • changedInput schema / properties / speaker_boost / description
      Previous value: -"Optional. Holds the voice closer to the speaker it was made from. Multilingual and flash only."New value: +"Optional. Holds the voice closer to the speaker it was made from. Every model except v4 and v4 turbo."
    • changedInput schema / properties / speed / description
      Previous value: -"Optional, 0.7 to 1.2. Under 1 is slower, over 1 is faster. Multilingual and flash only."New value: +"Optional, 0.7 to 1.2. Under 1 is slower, over 1 is faster. Every model except v4 and v4 turbo."
    • changedInput schema / properties / style / description
      Previous value: -"Optional, 0 to 1. Extra performance. Multilingual and flash only."New value: +"Optional, 0 to 1. Extra performance. Every model except v4 and v4 turbo."
    • changedInput schema / properties / text / description
      Previous value: -"The words to speak, exactly as they should be said, under 5000 characters. Write numbers and dates the way you want them read out."New value: +"The words to speak, exactly as they should be said. How long one take can be is set by the model you pick. Write numbers and dates the way you want them read out."
  4. Added

TDQS

A4.8/5.0
Behavior5/5

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

Beyond the annotations, the description discloses meaningful behavior: the model choices, 10,000-character v4 limit, lack of a stated turbo limit, price check before generation, billing receipt in the result, and word-level timing data. It also clarifies ignored parameters, which is unusually helpful for a mutation tool.

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?

The purpose and key constraints are front-loaded, and most sentences carry useful information. It is somewhat longer and more conversational than necessary, but the length is justified by the model nuances and output timing details.

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 generation tool with no output schema, the description explains what comes back: audio saved to the library, a billing receipt, and optionally word-by-word timestamps. Together with the model limits and sibling routing, this is complete enough for an agent to call it correctly.

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

Parameters4/5

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

Schema coverage is 100%, so the schema already carries baseline parameter meaning. The description adds useful semantics beyond that, especially the quality-to-model mapping and model-specific length constraints, though it does not cover every parameter such as language or request_id in the prose.

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?

The description states a specific verb and resource: speak written words aloud in an ElevenLabs voice and save the audio. It explicitly distinguishes this from sibling use cases such as generate_sound_effect and generate_dialogue, so an agent can identify its role without opening the 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?

It names concrete alternatives and the conditions that select them: use generate_sound_effect for sound effects and generate_dialogue when two or more people talk in one file. It also directs voice selection through list_voices, leaving little ambiguity about when to use this tool versus those siblings.

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

All switch toolsInspect
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