Developers

Build audio → data pipelines in minutes

Send audio files. Receive structured JSON.
Schema-defined, async, webhook-ready.

How developers use VoiceServant

Backend ingestion

Upload audio files from your app and receive structured JSON via webhook callbacks.

Automation pipelines

Process audio files automatically at scale.

Human-in-the-loop

Use Studio for template setup, API for production processing.

Template-first extraction

VoiceServant uses templates to define the structure of extracted data. Templates act as a contract between audio files and outputs.

1

Define schema (one-time setup)

Create a template in VoiceServant Studio. Upload a sample audio file and define how data should be extracted from the transcript. This is a one-time configuration step.

2

API processing

Send audio files to the API using the template ID. Jobs run asynchronously and return structured JSON via webhooks.

3

Structured output

Receive predictable, deterministic outputs that match your template schema.

What you get

Schema-defined extraction
Async job processing
Webhook callbacks
JSON, CSV, XLSX outputs
Batch processing
Deterministic templates
No training on customer data by default

API example

Request

POST /v1/documents
Authorization: Bearer sk_live_...

{
  "template_id": "audio_v1",
  "file_url": "https://example.com/meeting.mp3"
}

Response

{
  "job_id": "job_123",
  "status": "processing"
}

Webhook callback

{
  "status": "completed",
  "data": {
    "invoice_number": "INV-2048",
    "total": 1127.52,
    "currency": "USD"
  }
}

Security and data handling

Encrypted in transit
Configurable data retention
GDPR-friendly
No training on customer data by default

Get started

Read the API documentation and try it with your own audio files.