Voice Go API

All requests are made over HTTPS to the base URL below. Request bodies are JSON (multipart/form-data when uploading a file for transcription); the text-to-speech response is the audio file itself.

https://www.voicego.net/api/v1

Send your API key in the Authorization header with every request. Keys start with vg_live_. The X-API-Key header is also accepted.

Authorization: Bearer vg_live_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

POST/api/v1/tts

Voices the text with the chosen voice and returns the audio file. Each successful request deducts the text's character count from your monthly quota.

Parameters

curl -X POST https://www.voicego.net/api/v1/tts \
  -H "Authorization: Bearer $VOICEGO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Şirketimizi aradığınız için teşekkür ederiz.",
    "voice": "yeliz",
    "format": "wav_8k"
  }' \
  --output anons.wav

Response

A successful response is the audio file itself with status 200 (Content-Type: audio/mpeg or audio/wav). Response headers:

• X-Characters — Characters deducted from your quota for this request
• X-Quota-Limit · X-Quota-Used · X-Quota-Remaining — Monthly quota, usage this month and remaining characters

GET/api/v1/voices

Returns the voices you can use. HD voices with expressive: true support emotions, instructions and tags.

curl https://www.voicego.net/api/v1/voices -H "Authorization: Bearer $VOICEGO_API_KEY"

{
  "voices": [
    {
      "id": "yeliz",
      "name": "Yeliz",
      "names": { "en": "Amelia", "es": "Julia" },
      "gender": "female",
      "language": "tr-TR",
      "quality": "hd",
      "expressive": true
    }
  ]
}

POST/api/v1/transcribe

Transcribes an audio or video recording, separates speakers and gives the start and end time of each segment. Each successful request deducts the recording's duration (in seconds, rounded up) from your monthly transcription time.

You can send the file in two ways: upload files up to 4 MB directly as "file" with multipart/form-data; for larger files, pass a downloadable https address of the file as "url" in a JSON body (e.g. a signed, expiring link from your storage). Common audio and video formats such as MP3, WAV, M4A, OGG, FLAC, MP4, MOV and WEBM are supported.

Parameters

# 4 MB'a kadar dosya: doğrudan yükleme
curl -X POST https://www.voicego.net/api/v1/transcribe \
  -H "Authorization: Bearer $VOICEGO_API_KEY" \
  -F "file=@gorusme.mp3" \
  -F "language=tr" \
  -F "speakers=2"

# Büyük dosya: indirilebilir adres, SRT altyazı olarak
curl -X POST https://www.voicego.net/api/v1/transcribe \
  -H "Authorization: Bearer $VOICEGO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://depo.ornek.com/egitim-videosu.mp4",
    "language": "tr",
    "format": "srt"
  }' \
  --output altyazi.srt

Response

Response in json format; with txt and srt the response is the text file itself. Response headers:

• X-Billed-Seconds — Seconds deducted from your transcription time for this request
• X-Transcription-Limit-Seconds · X-Transcription-Used-Seconds · X-Transcription-Remaining-Seconds — Monthly transcription time, usage this month and remaining time (seconds)

{
  "text": "Merhaba, Voice Go'ya hoş geldiniz. Teşekkürler, sizi dinliyorum.",
  "language": "tur",
  "duration_seconds": 6.84,
  "billed_seconds": 7,
  "speakers": ["speaker_1", "speaker_2"],
  "segments": [
    { "speaker": "speaker_1", "start": 0.12, "end": 2.9, "text": "Merhaba, Voice Go'ya hoş geldiniz." },
    { "speaker": "speaker_2", "start": 3.4, "end": 6.7, "text": "Teşekkürler, sizi dinliyorum." }
  ]
}

A request is accepted as long as you have time remaining, and the whole recording is transcribed; if it is longer than your remaining time, the difference is also counted. Long recordings can take a few minutes; set your client timeout to at least 5 minutes.

GET/api/v1/usage

Returns your monthly character quota, transcription time, your usage and when the quota resets. Quotas are shared with the studio and reset on the 1st of each month (UTC).

{
  "characters": { "limit": 500000, "used": 128450, "remaining": 371550 },
  "transcription_seconds": { "limit": 36000, "used": 5420, "remaining": 30580 },
  "period": { "resets_at": "2026-11-01T00:00:00.000Z" }
}

With HD voices you can add tags inside the text for pauses, whispers, laughter and more. On voices that don't support them, tags are removed without being read.

"text": "Kampanyamız başladı! [pause] Detaylar için bire basın."

Errors are returned as JSON: { "error": { "code": "...", "message": "..." } }.

{
  "error": {
    "code": "quota_exceeded",
    "message": "Your monthly character quota has been used up."
  }
}

• Up to 2500 characters per request.
• Up to 60 text-to-speech and 10 transcription requests per minute per key. Exceeding it returns 429.
• Direct uploads for transcription are limited to 4 MB; send larger files with url.
• A monthly character quota and transcription time are assigned to your account; studio and API usage share the same quota.
• Contact your account manager to increase your quota.

• Use your API key only on the server side; never put it in browser or mobile app code.
• If you think your key has leaked, tell us; we'll revoke it immediately and issue a new one.
• Keys are stored only as encrypted hashes; nobody, including us, can see your key.
• The text and recordings you send are processed only to generate audio and transcripts, and are never used to train AI models.
• Recordings sent for transcription and their transcripts are not stored on Voice Go servers; the result is only returned to you.