funasr-zh-tw-mcp
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": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| transcribe_audioA | 將本機音檔轉成繁體中文逐字稿,支援 wav/mp3/m4a/flac/ogg/mp4 等格式。全程在本機運算,音檔不會上傳到任何雲端服務。注意:模型第一次載入約需 40 秒,若使用者在意等待時間可先呼叫 preload_model。 |
| preload_modelA | 預先把 FunASR 模型載入記憶體,讓之後的 transcribe_audio 呼叫可以立即開始辨識。首次執行會下載約 1.5GB 的模型權重。模型已載入時呼叫不會有任何副作用。 |
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 2 tools
transcribe_audio performs the core transcription while preload_model handles model initialization; their purposes are completely distinct and cannot be confused. The descriptions clearly separate the main task from an optional optimization step.
Both tools follow the same verb_noun pattern: transcribe_audio and preload_model. The verb clearly indicates the action, and the noun indicates the target, making the naming predictable and consistent.
Two tools is appropriate for a narrow-purpose server: one primary transcription tool plus an optional model preloading optimization. Each tool earns its place without redundancy, and adding more tools would likely be unnecessary for this focused scope.
The server's domain is local audio transcription, and the tool set covers the full workflow: optional model preloading followed by transcription. There are no obvious gaps for the stated purpose, and the preload_model tool explicitly handles the cold-start concern.