dataset/dev/1012.json

{"text": "A female speaker delivers a expressive and animated discourse with a significantly high-pitched voice. The recording is of acceptable quality, with a slight echo and only subtle background noise.", "speaker_id": "10644"}