GLM-ASR-2512

POST

/bigmodel/api/paas/v4/audio/transcriptions

GLM-ASR-2512 is Zhipu's new-generation speech recognition model, which supports the real-time conversion of speech into high-quality text. Whether it is daily chat, meeting minutes, work documents, or scenarios involving professional terms, it can accurately recognize and convert, greatly improving the efficiency of input and recording. The model maintains industry-leading recognition performance under multi-scenario and multi-accent conditions, with a Character Error Rate (CER) of only 0.0717, providing a fast and reliable speech input experience.

Price: 0.025 PTC/M token

Request

Header Params

Body Params application/json

Example

{
      "file": "https://file.302.ai/gpt/resource302db/20251215/f864885c6bcb40ec89789c11679ffbac.wav",
      "model": "glm-asr-2512"
}

Request Code Samples

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

curl --location --request POST 'https://api.302.ai/bigmodel/api/paas/v4/audio/transcriptions' \
--header 'Authorization: Bearer ' \
--header 'Content-Type: application/json' \
--data-raw '{
      "file": "https://file.302.ai/gpt/resource302db/20251215/f864885c6bcb40ec89789c11679ffbac.wav",
      "model": "glm-asr-2512"
}'

Responses

🟢200success

application/json

Body

Example

{
    "created": 1765790102,
    "id": "2025121517150122c05020dd894d4b",
    "request_id": "2025121517150122c05020dd894d4b",
    "text": "\nOkay, now please tell me how do you know from this picture that its location is Bangladesh?",
    "usage": {
        "completion_tokens": 21,
        "prompt_tokens": 189,
        "total_tokens": 210
    }
}

Modified at 2025-12-15 09:14:51

Wavespeed Retrieval Task

wavespeed-ai/infinitetalk-fast/video-to-video