The Balancing Act: Unmasking and Alleviating ASR Biases in Portuguese

In the field of spoken language understanding, systems like Whisper and Multilingual Massive Speech (MMS) have shown state-of-the-art performances. This study is dedicated to a comprehensive exploration of the Whisper and MMS systems, with a focus on assessing biases in automatic speech recognition (ASR) inherent to casual conversation speech specific to the Portuguese language. Our investigation encompasses various categories, including gender, age, skin tone color, and geo-location. Alongside traditional ASR evaluation metrics such as Word Error Rate (WER), we have …

act art asr automatic speech recognition balancing act biases conversation cs.ai cs.cl cs.cy exploration focus language language understanding massive mms multilingual performances recognition speech speech recognition spoken spoken language understanding state study systems understanding whisper