科学家正借助人工智能和机器学习技术推进动物交流研究。成立于2018年的地球物种项目（ESP）致力于解读动物之间的通讯方式，通过水听器、可穿戴式生物记录仪等设备采集数据，结合先进的算法分析包括草原犬鼠叫声和乌鸦鸣叫在内的各类动物声音1。研究团队开发的NatureLM模型已用人类语音、音乐和动物声音进行训练，使得研究人员能够检测腐食乌鸦的发声并区分不同鸟类的叫声1。
尽管研究已取得进展，但科学家对这一领域的局限性保持清醒认识。特拉维夫大学的神经生态学家Yossi Yovel警告称AI模型可能过度强调声学相似性而产生错误1。然而，研究者Con Slobodchikoff指出，"很多动物具有进化结构相似性，我们应该能够理解它们的交流内容"1。虽然实现动物与人类的完全双向对话仍面临挑战，但科学家认为深入理解动物交流将有助于揭示其内在思想与情感1。
Scientists are leveraging artificial intelligence and machine learning to unlock the secrets of how animals communicate with one another.1 The Earth Species Project, established in 2018, has been founded specifically to translate animal communication.1 Using hydrophones, wearable biological recorders, and advanced algorithms, researchers have made progress in detecting and analyzing animal vocalizations including prairie dog calls and crow cries.1
A key tool in this effort is NatureLM, a model developed by the Earth Species Project that has been trained on human speech, music, and animal sounds.1 Machine learning techniques have enabled researchers to identify vocalizations from carrion crows and distinguish between different bird species' calls.1 However, challenges remain in this emerging field. Yossi Yovel, a neuroecologist at Tel Aviv University, has cautioned that AI models risk making errors by overemphasizing acoustic similarities.1 Con Slobodchikoff has noted that many animals share evolutionary structural similarities, suggesting that understanding their communication should be achievable.1
While achieving true two-way dialogue between animals and humans remains distant and fraught with technical and conceptual obstacles, scientists believe that decoding animal communication could reveal insights into the inner thoughts and emotions of other species.1
评论
还没有评论，欢迎留下第一条。