GDPR-compliant Speech-to-Text API. 99 languages. Servers exclusively in Germany. Audio data is deleted immediately and irreversibly after processing.
No sales calls. No vendor lock-in. Get started instantly.
Not every provider is allowed to process your data. We are — and cheaper than most EU competitors.
If you transcribe audio with personal data — interviews, patient conversations, client calls, internal meetings — GDPR Art. 28 requires the data processor to be located within the EU. US providers like OpenAI, AWS or Google are subject to the US CLOUD Act and can be compelled to disclose data. Our servers are exclusively in Germany (Hetzner, Falkenstein/Nuremberg). Audio data is deleted immediately after transcription. No training, no storage, no third-party sharing. Data Processing Agreement (DPA) available on request.
REST API. Multipart upload. JSON response with text and word timestamps.
curl -X POST https://babelfish.kreislauftechnik.de/v1/speech-to-text \ -H "Authorization: Bearer YOUR_API_KEY" \ -F "file=@interview.mp3" \ -F "language=en"
{
"transcription": {
"text": "Hello, welcome to the interview...",
"duration_sec": 1847.3,
"language": "en",
"segments": [
{"start": 0.0, "end": 2.8, "text": "Hello,"},
{"start": 2.8, "end": 5.1, "text": "welcome to the interview..."}
]
},
"billing": {
"minutes_used": 30.8,
"remaining_minutes": 329.2,
"plan": "starter"
}
}
Everything you need for integration.
WAV, MP3, M4A, OGG, FLAC, MP4, AAC, WebM — anything ffmpeg can read. Max 100 MB per file.
POST /v1/speech-to-text — transcription
GET /v1/health — status
GET /v1/plans — pricing
GET /v1/account — usage
POST /v1/checkout — buy plan
API key in header:Authorization: Bearer YOUR_KEY
Get your key after purchase or via free plan.
Servers: Hetzner, Germany (Falkenstein/Nuremberg). Audio files are irreversibly deleted after transcription. No training. DPA on request.