Skip to main content
POST
/
v1
/
speech-to-text
# Example with audio file
curl -X POST https://api.woolball.xyz/v1/speech-to-text \
  -H "Authorization: Bearer <YOUR_API_KEY>" \
  -F "audio=@input.wav;type=audio/wav" \
  -F "model=onnx-community/whisper-large-v3-turbo_timestamped" \
  -F "outputLanguage=pt" \
  -F "returnTimestamps=true" \
  -F "webvtt=true"

# Example with video file (using audio field)
curl -X POST https://api.woolball.xyz/v1/speech-to-text \
  -H "Authorization: Bearer <YOUR_API_KEY>" \
  -F "audio=@input.mp4;type=video/mp4" \
  -F "model=onnx-community/whisper-large-v3-turbo_timestamped" \
  -F "outputLanguage=pt" \
  -F "returnTimestamps=true"

# Example with URL
curl -X POST https://api.woolball.xyz/v1/speech-to-text \
  -H "Authorization: Bearer <YOUR_API_KEY>" \
  -F "url=https://example.com/media.mp4" \
  -F "model=onnx-community/whisper-large-v3-turbo_timestamped" \
  -F "outputLanguage=pt"

{
  "data": {
    "text": "Transcribed text content",
    "chunks": [
      {
        "timestamp": [0, 2.5],
        "text": "First segment"
      },
      {
        "timestamp": [2.5, 5.0],
        "text": "Second segment"
      }
    ],
    "webvtt": "WEBVTT\n\n00:00:00.000 --> 00:00:02.500\nFirst segment\n\n00:00:02.500 --> 00:00:05.000\nSecond segment"
  }
}

Documentation Index

Fetch the complete documentation index at: https://docs.woolball.xyz/llms.txt

Use this file to discover all available pages before exploring further.

Convert audio or video files to text using speech recognition. Supports various audio and video formats.

​
Form Fields

​
audio
string
Audio or video file to transcribe (mutually exclusive with url)
​
url
string
URL to audio or video file to transcribe (mutually exclusive with audio) Example: “https://example.com/media.mp4”
​
model
string
Speech recognition model to use Example: “onnx-community/whisper-large-v3-turbo_timestamped”
​
outputLanguage
string
Target language for the transcription output Example: “pt”
​
returnTimestamps
boolean
Whether to return timestamps for each transcribed segment Example: true
​
webvtt
boolean
Generate WebVTT caption output (requires returnTimestamps=true) Example: true

​
Response

​
data
object
The transcription result object containing:
  • text: The transcribed text content
  • chunks: Array of segments with timestamps (when returnTimestamps=true)
  • webvtt: WebVTT formatted captions (when webvtt=true and returnTimestamps=true)

​
Status Codes

​
200
object
OK - Successful request
​
400
object
Bad Request - Validation error occurred
​
401
object
Unauthorized - Authentication failed
​
402
object
Payment Required
# Example with audio file
curl -X POST https://api.woolball.xyz/v1/speech-to-text \
  -H "Authorization: Bearer <YOUR_API_KEY>" \
  -F "audio=@input.wav;type=audio/wav" \
  -F "model=onnx-community/whisper-large-v3-turbo_timestamped" \
  -F "outputLanguage=pt" \
  -F "returnTimestamps=true" \
  -F "webvtt=true"

# Example with video file (using audio field)
curl -X POST https://api.woolball.xyz/v1/speech-to-text \
  -H "Authorization: Bearer <YOUR_API_KEY>" \
  -F "audio=@input.mp4;type=video/mp4" \
  -F "model=onnx-community/whisper-large-v3-turbo_timestamped" \
  -F "outputLanguage=pt" \
  -F "returnTimestamps=true"

# Example with URL
curl -X POST https://api.woolball.xyz/v1/speech-to-text \
  -H "Authorization: Bearer <YOUR_API_KEY>" \
  -F "url=https://example.com/media.mp4" \
  -F "model=onnx-community/whisper-large-v3-turbo_timestamped" \
  -F "outputLanguage=pt"

{
  "data": {
    "text": "Transcribed text content",
    "chunks": [
      {
        "timestamp": [0, 2.5],
        "text": "First segment"
      },
      {
        "timestamp": [2.5, 5.0],
        "text": "Second segment"
      }
    ],
    "webvtt": "WEBVTT\n\n00:00:00.000 --> 00:00:02.500\nFirst segment\n\n00:00:02.500 --> 00:00:05.000\nSecond segment"
  }
}