ai speech profile enroll
The ai speech profile enroll command enrolls a given voice profile using an audio sample.
Usage
Audio sample length requirements
For TextIndependentIdentification, the enrollment audio sample must contain 30 seconds or more of speech to be successful. For TextIndependentVerification, three or more samples with a combined 15 seconds or more of speech are required.
Options
| Option | Description |
|---|---|
--key KEY |
The subscription key to use. |
--region REGION |
The region for the speech resource. |
--id ID |
The ID of the voice profile to enroll. |
--file FILE |
The path to the audio sample file used for enrollment. |
--kind KIND |
The kind of profile enrollment (TextIndependentIdentification or TextIndependentVerification). |
--output file PATH |
The output file path to save the resulting data. |
--output json PATH |
The output JSON file path to save the resulting data. |