> ## Documentation Index
> Fetch the complete documentation index at: https://docs.api.ecylt.com/llms.txt
> Use this file to discover all available pages before exploring further.

# W-Speech (ASR)

> Call the free AI speech recognition (transcribe) provided by the WZH Team

<ParamField path="https://aispeech.api.ecylt.top" type="POST">
  Base API endpoint
</ParamField>

#### Query Parameters

Authorization: Bearer 114514\
Authentication: Add `Bearer 114514` in the Header

Body type: application/json

<ParamField body="audio_m" type="string" required>
  The format of the audio input. Can be either `base64` (base64 encoded string) or `wav_url` (URL of a .wav file)
</ParamField>

<ParamField body="audio" type="string" required>
  The audio data, either a base64 encoded string (if audio\_m is `base64`) or a URL to a .wav file (if audio\_m is `wav_url`)
</ParamField>

<ParamField body="language" type="string">
  The language of the audio. Default is `en`
</ParamField>

<ParamField body="vad_filter" type="string">
  Whether to use voice activity detection (VAD) filtering. Default is `false`
</ParamField>

<ResponseExample>
  ```json 200: OK normal response theme={null}
  {
    "transcription_info": {
      "duration": 10.5,
      "sample_rate": 16000,
      "num_words": 25
    },
    "text": "This is a sample transcription of the audio input.",
    "word_count": 7
  }
  ```
</ResponseExample>