PresentID Speaker verification API checks whether two voices belong to the same person or not. This capability is potentially useful in call centers.
We have proposed a deep learning-based method for speaker verification. Our team worked on this project for more than 1 year and the accuracy has passed over benchmarks such as the accuracy of the paper by Andrew Zisserman Group at Oxford University. In contrast with other methods that are text-dependent, our model is text and language-independent. On the other hand, the processing speed of our model is less than 1 sec and the model verifies a person by just two voices with a length of 4 secs. We have trained the model on tracks with English, French, Spanish, German, Persian, and Arabic languages. Our model is robust to the environment and virtual noises.
Youtube Videos
Input:
- Voice file
- Voice URL link
- Base64 Voice
Output:
- Result index
- Result message
Features:
- Accuracy over 90%.
- Less than 1 second processing time.
- No need for GPU.
- Language & text-independent.
- Easy integration with your app.
- Support IOS, Android, Windows and Mac devices.
- Easy integration with your app.
Use Cases:
- Call center
Rules & Restrictions:
- Send data via Base64 or a voice URL or voice file.
- The voice must be between three seconds and one minute.
- The voices must not exceed 5 MB.
- Supported file types: WAV, MP3, M4A, FLAC, AAC, OGG.