API Reference
One endpoint. All the controls. Transcribe anything.
1. Create API Key
Manage your keys in the User Panel after signing in. One click to generate, copy, and you're ready.
2. The Endpoint
/api/transcribeThis is the only endpoint you need. Send audio, get back perfect transcription with diarization, timestamps, and confidence scores.
3. Parameters
fileaudio/*, video/mp4requiredAudio file or MP4 video to transcribe (mp3, wav, m4a, flac, mp4)
num_speakersintegerExpected number of speakers (auto-detected if omitted)
custom_vocabularystringComma-separated domain terms for better accuracy
formatting_instructionstringStyle guide: bullet points, paragraphs, no timestamps, etc. Requires enhance_with_ai to be true and a PRO plan.
enhance_with_aibooleanSet to 'false' for Strict Privacy Mode (bypasses OpenAI). Defaults to true for PRO users.
languagestringLanguage code (e.g., 'pl', 'en'). Auto-detected if omitted or set to 'auto'.
show_speaker_infobooleanInclude speaker labels in the text output. Defaults to true.
show_timestampsbooleanInclude timestamps in the text output. Defaults to true.
4. Limits
File Limits: Maximum file size is 1000 MB. Maximum audio duration is 300 minutes (5 hours) per request. Your account must have sufficient API Credits (PAYG) to process the file length.
Try It
import requests
url = "https://voxora.me/api/transcribe"
headers = {"Authorization": "Bearer YOUR_API_KEY"}
files = {"file": ("audio.mp3", open("audio.mp3", "rb"), "audio/mpeg")}
data = {
"num_speakers": "2",
"custom_vocabulary": "specialized terms here",
"formatting_instruction": "Translate to Spanish, remove all punctuation",
"enhance_with_ai": "true",
"language": "en",
"show_speaker_info": "true",
"show_timestamps": "true",
}
resp = requests.post(url, headers=headers, files=files, data=data)
print(resp.json())Response: JSON with text, speakers, timestamps, and confidence.