HomeVoice Detection › API

Voice detection API — what actually exists.

Short version: there is no public voice endpoint on the TextSight API today. The public API covers text — detection, scoring and rewriting — and voice detection currently runs only through the web tool. Rather than publish a page implying otherwise and collect sign-ups for something that does not exist, here is exactly what is available, what is not, and how to reach us if you have a volume use case.

Read the API docs Talk to us about volume
Text API live No voice endpoint yet
Status

What is available today.

Plainly, so you can decide in thirty seconds whether to keep reading.

✓ Available on the public API

  • POST /v2/detect — AI detection on text, with per-sentence scores
  • POST /v2/score — the same engine, scores only, lighter payload
  • POST /v2/rewrite — humanize text

× Not available on the public API

  • Any audio or voice endpoint
  • Bulk or async audio job submission
  • Engine attribution — the web tool does not return it either
  • Per-segment timelines or forensic reports

Voice detection is available through the web tool: upload MP3, WAV, M4A, OGG or FLAC up to 10 MB and get an AI-likelihood score and a human-versus-synthetic verdict. Three checks a day, free, whether or not you are signed in. Voice has its own plans, separate from our text subscriptions — Echo $9/mo for 100 checks a month, Amplify $29 for 1,000, Broadcast $99 for 10,000 — plus one-time Soundbites credit packs from $5. That is a manual workflow, not something you can put a queue through.

Why say this so bluntly? Because the alternative is a page that ranks for “voice detection API”, collects your email, and wastes a week of your evaluation time. If voice at volume is a hard requirement for you right now, you should know that in the first paragraph.
What exists

The text API, concretely.

If text detection is also useful to you, this part is live and documented.

Base URL is https://api.textsight.ai/v2. Authenticate with an API key — either an Authorization: Bearer header or an x-api-key header. Live keys are prefixed sk_live_.

# AI detection on text — returns overall + per-sentence scores
curl -X POST https://api.textsight.ai/v2/detect \
  -H "Authorization: Bearer sk_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"text": "The text you want to check."}'

The response carries a humanization_score, an ai_probability, a verdict, a confidence value, per-sentence breakdown and a request_id. /v2/score returns the same numbers without the sentence array when you do not need it.

Scopes are enforced per key: detection and scoring need the scan scope, rewriting needs humanize. Quota is counted in API calls per month and varies by plan. Full parameters, error codes and rate limits are in the API documentation.

To be unambiguous: none of these endpoints accept audio. Sending an audio file to /v2/detect will not work — it takes text.
If you need voice

Talk to us, and tell us the specifics.

What would actually help us give you a straight answer.

If you have a genuine volume use case for voice detection — a fraud queue, a moderation pipeline, a research corpus — get in touch. We would rather have a concrete conversation than a waitlist. The things that let us respond usefully rather than vaguely:

  • Volume and shape. Files per day, and whether that is a steady trickle or occasional large batches.
  • Audio characteristics. Where it comes from — telephony, uploads, messaging apps — plus typical duration and format. This matters more than volume, because telephony audio is the hardest case in the category and it changes what is honestly achievable.
  • Latency tolerance. Whether an asynchronous job with a callback works, or you need a synchronous response.
  • What the result feeds. A human reviewer, an automated decision, or a research dataset. This is the important one: we would want to talk you out of an automated decision, because a detector output is not a sound basis for one.
  • Compliance constraints. Data residency, retention requirements, whether recordings can leave your environment at all.
A note on what we will not build for you. Real-time detection on a live call is not something we offer, and we would be sceptical of anyone who claims to. Detection runs on recorded files. If your requirement is live protection on a phone line, the honest answer is that process controls — out-of-band callback and the rest — will serve you better than any vendor's roadmap.
Not forensic-grade. A TextSight voice result is a triage signal, not proof, and it must never be the sole basis for a disciplinary, employment, financial or legal decision. Our full position on what a result is worth is here.
FAQ

API questions.

Is there a TextSight voice detection API?
Not publicly, today. The public API covers text detection, scoring and rewriting. Voice detection is available through the web tool only. If you have a volume use case, contact us with the specifics and we will give you a straight answer rather than a waitlist.
Can I send audio to the /v2/detect endpoint?
No. It accepts text. There is no audio or voice endpoint on the public API at present.
What does the public API actually cover?
POST /v2/detect for AI detection on text with per-sentence scores, POST /v2/score for the same scores with a lighter payload, and POST /v2/rewrite to humanize text. Base URL https://api.textsight.ai/v2, authenticated with a Bearer or x-api-key header using a key prefixed sk_live_.
Is there an async or bulk endpoint for audio?
No. Neither bulk submission nor asynchronous audio jobs exist on the public API today.
Can the API tell me which AI tool generated a voice?
No, and neither can the web tool. Engine attribution is not something our detector returns. We do not display a guess we cannot stand behind.
Do you offer real-time detection on live calls?
No. Detection runs on recorded files, not live streams, and we would treat claims of consumer-scale live detection with scepticism. For live protection on a phone line, process controls such as out-of-band callback are more effective than any detection product.
Related

More voice detection guides.

Need voice at volume? Tell us what you're building.

Three checks a day, free, no signup. Your audio is never stored.

Try the web tool All voice guides
Public API is text-only today · voice detection runs through the web tool