| dataset_info: | |
| features: | |
| - name: audiocap_id | |
| dtype: string | |
| - name: youtube_id | |
| dtype: string | |
| - name: start_time | |
| dtype: string | |
| - name: audio | |
| dtype: | |
| audio: | |
| sampling_rate: 48000 | |
| - name: caption | |
| dtype: string | |
| splits: | |
| - name: train | |
| - name: validation | |
| - name: test | |
| license: mit | |
| task_categories: | |
| - text-to-speech | |
| language: | |
| - en | |
| multilinguality: | |
| - monolingual | |
| tags: | |
| - youtube | |
| - captions | |
| pretty_name: AudioCaps | |
| size_categories: | |
| - 10K<n<100K | |
| source_datasets: | |
| - original | |
| paperswithcode_id: audiocaps | |