Skip to main content
Glama

APICK Convert

by app.apick
OverviewDiscussionsAdmin

Gemini TTS 작업 접수

tts_gemini_create

Gemini 3.8 Flash-Lite TTS에 목소리·낭독 스타일·본문 또는 화자별 발화를 전달합니다. 기본 on 정규화는 합계 2,000자 한도와 스킬 요금이 적용됩니다. 꺼도 MP3·원문 ASS 결과 구조는 같습니다. [동기화된 실제 토큰 원가×환율×1.4(작업당 기본요금 5P, 2026-11-06부터 · 소수점 올림) + 정규화 스킬 요금]

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
paceNo말 속도 배율 0.5~2.0. 1이 기본
textNoutterances와 둘 중 하나
toneNo용도·톤. 생략 시 기본
pitchNo음높이 -12~12 반음. 0이 기본
styleNo자유 서술 낭독 스타일
accentNo억양. 생략 시 표준어
emotionNo감정 표현. 생략 시 기본
speakersNo화자 이름 → {voice_id, style, emotion, tone, accent, pace, pitch, volume_gain_db}
voice_idNo목소리 목록의 ID. 기본 Kore
utterancesNo각 항목 text, voice_id, style, speaker, emotion, tone, accent, pace, pitch, volume_gain_db
multi_speakerNo멀티화자. Gemini는 화자 2명까지 한 호출, ChatGPT는 발화별 목소리 대화
normalize_textNo기본 true, 정규화 추가 과금
volume_gain_dbNo음량 보정 -12~12 dB. 0이 기본
idempotency_keyNo같은 요청의 재접수 키

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedInput schema / properties / style / description
      Previous value: -"낭독 스타일"New value: +"자유 서술 낭독 스타일"
  2. Changed9 schema fields changed
    • addedInput schema / properties / accent
      Added value: +{
+  "description": "억양. 생략 시 표준어",
+  "enum": [
+    "standard",
+    "seoul",
+    "gyeongsang",
+    "jeolla",
+    "chungcheong"
+  ],
+  "type": "string"
+}
    • addedInput schema / properties / emotion
      Added value: +{
+  "description": "감정 표현. 생략 시 기본",
+  "enum": [
+    "neutral",
+    "calm",
+    "cheerful",
+    "excited",
+    "sad",
+    "serious",
+    "friendly",
+    "empathetic",
+    "confident",
+    "gentle",
+    "whisper",
+    "narration"
+  ],
+  "type": "string"
+}
    • addedInput schema / properties / multi_speaker
      Added value: +{
+  "description": "멀티화자. Gemini는 화자 2명까지 한 호출, ChatGPT는 발화별 목소리 대화",
+  "type": "boolean"
+}
    • addedInput schema / properties / pace
      Added value: +{
+  "description": "말 속도 배율 0.5~2.0. 1이 기본",
+  "type": "number"
+}
    • addedInput schema / properties / pitch
      Added value: +{
+  "description": "음높이 -12~12 반음. 0이 기본",
+  "type": "number"
+}
    • addedInput schema / properties / speakers
      Added value: +{
+  "additionalProperties": {},
+  "description": "화자 이름 → {voice_id, style, emotion, tone, accent, pace, pitch, volume_gain_db}",
+  "propertyNames": {
+    "type": "string"
+  },
+  "type": "object"
+}
    • addedInput schema / properties / tone
      Added value: +{
+  "description": "용도·톤. 생략 시 기본",
+  "enum": [
+    "narration",
+    "news",
+    "audiobook",
+    "documentary",
+    "ad",
+    "conversation",
+    "announcement",
+    "tutorial",
+    "storytelling"
+  ],
+  "type": "string"
+}
    • changedInput schema / properties / utterances / description
      Previous value: -"각 항목 text, voice_id, style, speaker"New value: +"각 항목 text, voice_id, style, speaker, emotion, tone, accent, pace, pitch, volume_gain_db"
    • addedInput schema / properties / volume_gain_db
      Added value: +{
+  "description": "음량 보정 -12~12 dB. 0이 기본",
+  "type": "number"
+}
  3. Added

TDQS

B3.3/5.0
Behavior4/5

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

Annotations only declare the coarse profile (not read-only, not destructive, not idempotent, open-world). The description adds material context the annotations do not: normalization is on by default with a 2,000-character cap and a skill fee, and disabling it does not change the MP3/ASS output structure. That is genuinely additive, though it is silent on whether this is an async job that must be polled.

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

Conciseness3/5

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

The purpose sentence is front-loaded and efficient, but the trailing bracket is a dense, run-on pricing formula (token cost x FX x 1.4 + skill fee) that is hard to parse and partially duplicates the normalize_text fee already stated earlier.

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

Completeness3/5

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

For a 14-parameter, nested-object tool with no output schema, the description covers cost and normalization limits but omits the return contract: it never says the call produces a job handle consumed by tts_jobs_status/tts_jobs_result, which is the key thing an agent needs to chain the workflow.

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% across 14 parameters, so the schema already carries the per-parameter detail and the baseline is 3. The description adds only a little beyond it (the 2,000-character normalization cap and the text-vs-utterances alternative), so it does not rise above baseline.

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

Purpose4/5

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

The description names a specific verb (전달/submit voice, style, text or utterances) and a specific resource (Gemini 3.8 Flash-Lite TTS), which separates it from the tts_openai_create sibling. The purpose is clear despite the pricing bracket adding noise at the end.

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

Usage Guidelines2/5

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

There is no statement of when to pick this over tts_openai_create, tts_quote, or the tts_jobs_* siblings. The only conditional guidance is about the normalize_text default and its fee, which is parameter behavior rather than tool selection guidance.

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

All APICK Convert 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