dataset/dev/1001.json

{"text": "A male speaker with an American accent delivers slightly expressive and animated speech, slightly faster than a moderate pace. The recording has a slightly noisy background and a subtle roominess to the sound.", "speaker_id": "11988"}