dataset/dev/10.json

{"text": "A female English speaker delivers a very expressive and animated speech with a moderate pace and moderate pitch, but the recording quality is poor with a significant amount of background noise making the sound slightly close.", "speaker_id": "4627"}