Verification Models
Verification models are analyzers and comparators of the characteristics of the voice. They can be used to perform verifications.
Version selection: the platform automatically selects the latest versions of verification models.
Identity Verification Models
Identity verification models are used to analyze the level of similarity between two voices, comparing a new voice sample with a previoulsy created voiceprint. They can be used to perform identity verifications.
Model selection: the platform automatically selects appropriate identity verification models for the chosen voiceprints. Every voiceprint model has a matching identity verification model.
Identity Model Descriptions
| Model Name | Handle | Version | Status | Description |
|---|---|---|---|---|
| Digital | digital | v1.1 | ✅ Available | The digital identity verification model is specifically designed and trained for managing higher quality audio, with sample rates of 16KHz or higher. The English and Spanish languages make up a significant part of its training dataset. |
| Phone | phone | v1.1 | ✅ Available | The phone identity verification model is specifically designed and trained for managing phone audio, with sample rates of 8KHz or higher. The telephone channel and the English and Spanish languages make up a significant part of its training dataset. |
| Default | default | v1 | ⚠️ Deprecated | The default identity verification model is deprecated and will be removed in a future platform release. |
Identity Model Requirements
| Model Name | Audio Max Duration | Audio Min Voice Duration | Audio Min Frequency | Audio File State | Voiceprint Model | Voiceprint Model Version | Voiceprint State |
|---|---|---|---|---|---|---|---|
| Digital | 2 minutes | 1.5 seconds | 16000 Hz | Available | digital | v1 | Computed |
| Phone | 2 minutes | 1.5 seconds | 8000 Hz | Available | phone | v1 | Computed |
| Default | 2 minutes | 1.5 seconds | 16000 Hz | Available | default | v1 | Computed |
Authenticity Verification Models
Authenticity verification models are analyzers of the authenticity and liveliness of the voice. They can be used to perform authenticity verifications.
Authenticity Model Descriptions
| Model Name | Handle | Version | Status | Description |
|---|---|---|---|---|
| Digital | digital | v2 | ✅ Available | The digital authenticity verification model aims to determine if the voice is real or generated/modified by AI (deepfake). It is the recommended model for everything but the telephone channel. The digital channel and the English and Spanish languages make up a significant part of its training dataset. |
| Phone | phone | v2 | ✅ Available | The phone authenticity verification model aims to determine if the voice is real or generated/modified by AI (deepfake). It is the recommended model for the telephone channel and is the only model that supports audios of 8 KHz. The English and Spanish languages make up a significant part of its training dataset. |
| Default | default | v10 | ⚠️ Deprecated | The default authenticity verification model is deprecated and will be removed in a future platform release. |
| Telephone | telephone | v6 | ⚠️ Deprecated | The telephone authenticity verification model is deprecated and will be removed in a future platform release. |
Authenticity Model Requirements
| Model Name | Chunk Duration | Audio Max Duration | Audio Min Frequency | File State | Audio Chunks Minimum Voice Duration |
|---|---|---|---|---|---|
| Digital | 4 seconds | 5 minutes | 16000 Hz | Available | 750 milliseconds |
| Phone | 4 seconds | 5 minutes | 8000 Hz | Available | 750 milliseconds |
| Default | 4 seconds | 5 minutes | 16000 Hz | Available | 750 milliseconds |
| Telephone | 4 seconds | 5 minutes | 8000 Hz | Available | 750 milliseconds |
Message Verification Models
Message verification models are used to compare the content of a voice sample with an expected message to verify if the two of them match. They can be used as a voice captcha. They can be used to perform message verifications.
Message Model Descriptions
| Model Name | Handle | Version | Status | Description |
|---|---|---|---|---|
| Digital | digital | v1 | ✅ Available | The digital message verification model is specifically designed and trained for managing higher quality audio, with sample rates of 16KHz or higher.It supports a wide range of languages and expects a single phrase. |
| Phone | phone | v1 | ✅ Available | The phone message verification model is the recommended model for the telephone channel and is the only model that supports audios of 8 KHz.It supports a wide range of languages and expects a single phrase. |
| Default | default | v2 | ⚠️ Deprecated | The default message verification model is deprecated and will be removed in a future platform release. |
Message Model Requirements
| Model Name | Audio Max Duration | Audio Minimum Voice Duration | Audio Min Frequency | File State | Message Maximum Length | Supports Digits | Language |
|---|---|---|---|---|---|---|---|
| Digital | 30 seconds | 2.5 seconds | 16000 Hz | Available | 200 characters | No | All languages supported |
| Phone | 30 seconds | 2.5 seconds | 8000 Hz | Available | 200 characters | No | All languages supported |
| Default | 30 seconds | 2.5 seconds | 16000 Hz | Available | 200 characters | No | All languages supported |
Supported Languages
Must be one of the following:
en, zh, de, es, ru, ko, fr, ja, pt, tr, pl, ca,
nl, ar, sv, it, id, hi, fi, vi, he, uk, el, ms,
cs, ro, da, hu, ta, no, th, ur, hr, bg, lt, la,
mi, ml, cy, sk, te, fa, lv, bn, sr, az, sl, kn,
et, mk, br, eu, is, hy, ne, mn, bs, kk, sq, sw,
gl, mr, pa, si, km, sn, yo, so, af, oc, ka, be,
tg, sd, gu, am, yi, lo, uz, fo, ht, ps, tk, nn,
mt, sa, lb, my, bo, tl, mg, as, tt, haw, ln, ha,
ba, jw, su or yue.