leemysw/yovoice
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| logging | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| cancel_generationB | 按 requestId 取消任务；取消后查询至终态，终态任务保持不变 |
| generateA | 同步生成，长耗时优先使用 submit_generation。完整生成语音，返回 id、duration 和音频位置。path 为本机 WAV 文件的绝对路径 |
| get_generationA | 按 requestId 查询任务；建议每2–5秒查询，completed 后用 downloadPath 下载（本机模式直接返回 path） |
| list_modelsA | 查询模型 ID、安装状态和高级参数定义；生成前选择已安装模型 |
| list_voicesA | 列出已上传参考音色及 voiceId |
| render_scoreA | 把编曲乐谱渲染为混音 WAV，返回 id、duration 和音频位置（远程为 downloadPath，本机为 path）。乐谱以小节和拍为单位：tempo、timeSignature、可选 sections（段落及变速），tracks 每个声部含 program（GM 音色号）或 drums、role（melody/piano/strings/bass/drums/pad/arp/other，决定默认电平）、notes（bar、beat 从 1 开始、pitch 为 MIDI 音高、length 以拍计、velocity），可选 level、pan、reverb、humanize、dynamics。soundFont 选择已安装的音色库（musescore-general-sf2 默认、generaluser-gs-sf2、sgm-v2-sf2） |
| statusB | 查看推理引擎及当前任务状态 |
| submit_generationA | 提交异步语音生成，立即返回任务状态。requestId 为客户端生成的32位十六进制随机ID；相同ID和参数重试不会重复生成。随后用 get_generation 查询，完成后取得音频。优先使用此工具避免长连接超时 |
| upload_voiceA | 导入参考音频，返回 voiceId。传 path（本机音频文件路径，常见格式自动转换）或 Base64 编码的 audio，限 20 MB、1–60 秒 |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 9 tools
Most tools target clearly different resources or actions (upload_voice, list_voices, list_models, render_score, cancel_generation are all distinct). The only real overlap is generate vs submit_generation (sync vs async) and status vs get_generation, but the descriptions explicitly explain the difference and which to prefer, so an agent can choose correctly.
Seven of nine names follow a clean verb_noun pattern (list_voices, list_models, get_generation, submit_generation, cancel_generation, upload_voice, render_score). The bare noun 'status' and bare verb 'generate' are the only deviations, and both remain readable.
Nine tools is well within a comfortable range and every tool maps to a distinct need: voice import/listing, model discovery, engine status, sync/async generation, polling, cancel, and score rendering. Nothing feels padded or missing at the count level.
The voice lifecycle (upload → list → generate → poll → cancel) plus model discovery and score rendering is covered end to end. Minor gaps exist: no delete/replace voice, and no way to list past generations without knowing a requestId, but agents can work around these.