top of page

The outsourcing of multimedia content such as audio and speech data to Cloud Data Centers (CDCs) for storage and computation is becoming increasingly common due to their high storage and computational needs. Companies constrained in resources tend to benefit from the storage, high-end processing, elasticity, scalability and cost effective model provided by CDCs. However, the use of third party servers such as CDC raises security concerns due to the sensitive nature of audio and speech data. Data encryption is a commonly practiced solution to improve security. However, to process the data at CDCs, data often need to be decrypted, which also raising issues in security. Thus there is a need to protect the security of audio and speech records for storage and computation at CDC such that the CDC cannot learn any information about the confidential data.

This research focuses on the secure outsourcing of storage and computation of audio/speech to a CDC. Our methods are based on signal processing in encrypted domain (SPED) and (K,N) Shamir secret sharing. We propose schemes for:

 

 

1. Secure Cloud-based Audio Storage

2. Secure Cloud-based Audio Reverberation

3. Secure Cloud-based Speech Noise Reduction

bottom of page