Voice security analysis

Detect synthetic (deepfake) voices and verify liveness.

This endpoint requires authentication. Include your API key in the Authorization header. This is an async job-based endpoint. Results are typically available immediately after submission.

How It Works

Submit audio via POST /v1/voice-security — returns a job_id (typically already completed)
Retrieve results via GET /v1/voice-security/{job_id}/result — returns security signals

Request Body (multipart/form-data)

Parameter	Type	Required	Description
`file`	binary	Yes	Audio file (WAV, MP3, OGG, FLAC, WebM)
`model`	`valsea-voice-security`	Yes
`response_format`	`json` \| `verbose_json`	No	Default: `json`

Code Examples

curl -X POST https://api.valsea.ai/v1/voice-security \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "model=valsea-voice-security" \
  -F "file=@audio.wav"

Response

Submit Response (202)

Field	Type	Description
`job_id`	string	Unique job identifier
`status`	string	`completed` (results ready)

Minimal (`json`)

Field	Type	Description
`job_id`	string
`status`	string	`completed`
`synthetic_probability`	number	Probability the voice is synthetic (0-1)
`liveness_status`	`verified` \| `failed` \| `scanning`	Voice liveness verification

Verbose (`verbose_json`)

Same fields as the minimal response.

Security Signals

Signal	Range	Interpretation
`synthetic_probability`	0-1	< 0.5 = likely human, 0.5-0.7 = uncertain, > 0.7 = likely synthetic
`liveness_status`	enum	`verified` = human, `scanning` = uncertain, `failed` = synthetic suspect

Errors

Status	Description
`401`	Missing or invalid API key
`402`	Insufficient credits
`404`	Job not found or expired
`503`	Service not configured