@misc{indiciaeb0b2e6bc140c,
  title = {ELSA: Acoustic Event-Level Semantic Alignment for Fine-Grained Reference-Free Text-to-Audio Evaluation},
  author = {Shuntaro Suzuki and Kento Tokura and Daichi Yashima and Kanon Amemiya and Komei Sugiura and Shinnosuke Takamichi},
  year = {2026},
  url = {https://arxiv.org/abs/2606.17404},
  note = {Source identifier: 2606.17404}
}