Alibaba Cloud uses a wealth of corpus data to train speech recognition models for scenarios including general, educational, judicial, and medical scenarios. This allows Intelligent Speech Interaction to provide high-accuracy models for these scenarios. If these basic models do not meet your business requirements or if you want to optimize these basic models, you can use the self-learning platform to optimize these basic models.
You can use the self-learning platform to upload a training corpus. You can then select a basic model in a specific field and train the model based on the training corpus. This can improve the speech recognition rate of the specific scenario. This method is useful to optimize the recognition rate of specialized vocabularies and high-frequency words.
No additional charges are incurred for this feature regardless of whether you use Intelligent Speech Interaction Trial Edition or Intelligent Speech Interaction Commercial Edition.
Prepare a training corpus
Prepare a training corpus based on the call request limits and the optimization suggestions. For more information about the training corpus, see Introduction.
Train a custom model
For more information about how to create and apply custom models, see Create a custom linguistic model.
Manage a model
You can perform automated testing, retrain a model, and delete a model.
