Conecto

Transcribe audio

speech_transcribe

Read-onlyIdempotent

Turn the speech in an audio or video file from the file store (MP3, WAV, M4A, OGG, FLAC, WebM, MP4, MOV, up to 25 MB) into text, in any of about 100 languages, detected by itself. With speakers, it says who spoke when; with timestamps, it gives timed segments. Runs in the background: poll the call_id. Charged per started minute of audio, never more than max_minutes. Costs $0.01 per started minute of audio; nothing if none are found. Runs in the background: the reply has a call_id; poll it with the account_get_call tool or GET /v1/calls/.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "file_id": "<file_id>"
    },
    "name": "speech_transcribe"
  }
}

Input Schema

Name	Required	Description
`file_id`	Yes
`language`	No	ISO 639 code like "en" or "deu" when you know it.
`speakers`	No
`timestamps`	No
`max_minutes`	No	The most minutes to pay for; audio past it is transcribed but not charged.

Schema Changelog

Changes observed during successful MCP inspections.

First observedOct 28, 2026

TDQS

A4.6/5.0

Behavior5/5

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

Goes well beyond the readOnly/idempotent annotations: it discloses background execution, the polling requirement with a concrete sibling (account_get_call) and endpoint, an explicit cost model ($0.01/started minute, capped by max_minutes, free when nothing found), and size/format constraints.

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?

Front-loaded with the core action and constraints, but the background/polling instruction is stated twice ('poll the call_id' and again 'the reply has a call_id; poll it with...'), which is redundant and could be consolidated.

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

Completeness4/5

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

With no output schema, the description does explain the return shape (text, speaker attribution, timed segments, call_id). Only minor gaps remain, such as how the final transcript text is retrieved after polling.

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

Parameters5/5

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

With only 40% schema coverage, the description compensates by explaining what speakers ('says who spoke when'), timestamps ('timed segments'), language auto-detection, and max_minutes (cost cap) actually do. It meaningfully extends the schema rather than restating it.

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?

States a specific verb+resource ('turn the speech in an audio or video file ... into text') and even enumerates supported formats and the 25 MB limit. An agent can distinguish this from speech_synthesize (opposite direction) without opening any schema.

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

Usage Guidelines4/5

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

Explains the source ('from the file store'), the size ceiling, and the async workflow (poll the call_id). It does not explicitly contrast with siblings like speech_synthesize, but the direction of the operation makes the choice unambiguous.

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

All Conecto tools Inspect

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

Need Help?