CA-SSLR: Condition-Aware Self-Supervised Learning Representation for Generalized Speech Processing