Skip to main content
POST /v1/audio/transcriptions
Transcribes audio into text in the language of the audio.

​
Request Body

This endpoint accepts multipart/form-data.
​
file
file
required
The audio file to transcribe. Supported formats: flac, mp3, mp4, mpeg, mpga, m4a, ogg, wav, webm. Maximum file size is 25MB.
​
model
string
required
The model to use for transcription (e.g., whisper-1).
​
language
string
The language of the audio in ISO-639-1 format (e.g., en, es, fr). Providing the language improves accuracy.
​
prompt
string
Optional text to guide the model’s style or continue a previous transcript. Should match the audio language.
​
response_format
string
default:"json"
The output format. Options: json, text, srt, verbose_json, vtt.
​
temperature
number
default:"0"
Sampling temperature between 0 and 1. Higher values make output more random.

​
Response

Varies based on response_format:

​
JSON Response (default)

​
text
string
The transcribed text.

​
Verbose JSON Response

​
task
string
The task performed (transcribe).
​
language
string
The detected language.
​
duration
number
Duration of the audio in seconds.
​
text
string
The transcribed text.
​
segments
array
Array of transcript segments with timestamps.

​
Examples

​
Basic Transcription

from openai import OpenAI

client = OpenAI(
    api_key="sk-voidai-your_key_here",
    base_url="https://api.voidai.app/v1"
)

with open("audio.mp3", "rb") as audio_file:
    transcript = client.audio.transcriptions.create(
        model="whisper-1",
        file=audio_file
    )

print(transcript.text)

​
With Language Hint

with open("spanish_audio.mp3", "rb") as audio_file:
    transcript = client.audio.transcriptions.create(
        model="whisper-1",
        file=audio_file,
        language="es"
    )

​
SRT Subtitles

with open("video_audio.mp3", "rb") as audio_file:
    transcript = client.audio.transcriptions.create(
        model="whisper-1",
        file=audio_file,
        response_format="srt"
    )

# Save as subtitle file
with open("subtitles.srt", "w") as f:
    f.write(transcript)

​
Verbose JSON with Timestamps

with open("audio.mp3", "rb") as audio_file:
    transcript = client.audio.transcriptions.create(
        model="whisper-1",
        file=audio_file,
        response_format="verbose_json"
    )

print(f"Language: {transcript.language}")
print(f"Duration: {transcript.duration}s")
for segment in transcript.segments:
    print(f"[{segment['start']:.2f}s - {segment['end']:.2f}s] {segment['text']}")

​
Response Examples

​
JSON Response

{
  "text": "Hello, this is a test transcription of an audio file."
}

​
Verbose JSON Response

{
  "task": "transcribe",
  "language": "english",
  "duration": 5.42,
  "text": "Hello, this is a test transcription of an audio file.",
  "segments": [
    {
      "id": 0,
      "start": 0.0,
      "end": 2.5,
      "text": "Hello, this is a test",
      "tokens": [50364, 2425, 11, 341, 307, 257, 1500],
      "temperature": 0.0,
      "avg_logprob": -0.25,
      "compression_ratio": 1.2,
      "no_speech_prob": 0.01
    },
    {
      "id": 1,
      "start": 2.5,
      "end": 5.42,
      "text": " transcription of an audio file.",
      "tokens": [50489, 1112, 11, 295, 364, 6279, 2058],
      "temperature": 0.0,
      "avg_logprob": -0.22,
      "compression_ratio": 1.1,
      "no_speech_prob": 0.02
    }
  ]
}

​
SRT Response

1
00:00:00,000 --> 00:00:02,500
Hello, this is a test

2
00:00:02,500 --> 00:00:05,420
transcription of an audio file.

​
Tips

Providing the language parameter improves accuracy, especially for non-English audio or audio with accents.
The prompt parameter can help with proper nouns, technical terms, or specific formatting expectations.
Use srt or vtt for subtitles, verbose_json when you need timestamps, or plain text for simple transcripts.