Using fine-tuning for the embedding model

Hello everyone,
Is there any way to use fine-tuning for our similarity embedding model (text-similarity-ada-001) to generate a new embedding?