Datasets:

Modalities:
Audio
Text
Formats:
parquet
Languages:
English
ArXiv:
Libraries:
Datasets
Dask
License:
File size: 2,341 Bytes
16f7bc7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4c8d269
16f7bc7
4c8d269
 
16f7bc7
 
 
00c755e
4c8d269
00c755e
 
 
 
 
 
 
 
 
 
 
 
 
 
d01715e
 
 
16f7bc7
86d81b5
 
 
 
b9c0c30
d01715e
86d81b5
 
 
 
bbe04aa
86d81b5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
d01715e
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: correct_text
    dtype: string
  - name: correct_audio
    dtype: audio
  - name: negative_text
    dtype: string
  - name: negative_audio
    dtype: audio
  splits:
  - name: train
    num_bytes: 2660941782.232
    num_examples: 1871
  download_size: 2439350149
  dataset_size: 2660941782.232
configs:
- config_name: default
  data_files:
  - split: tsc_bm
    path: tSC/bm/train-*
  - split: tsc_am
    path: tSC/am/train-*
  - split: tsc_bf
    path: tSC/bf/train-*
  - split: tsc_af
    path: tSC/af/train-*
  - split: ssc_bm
    path: sSC/bm/train-*
  - split: ssc_am
    path: sSC/am/train-*
  - split: ssc_bf
    path: sSC/bf/train-*
  - split: ssc_af
    path: sSC/af/train-*
license: mit
language:
- en
---


# Multi Speaker StoryCloze

A multispeaker spoken version of [StoryCloze](https://paperswithcode.com/dataset/storycloze) Synthesized with [Kokoro TTS](https://huggingface.co/hexgrad/Kokoro-82M).
The dataset was synthesized to evaluate the performance of speech language models as detailed in the paper ["Scaling Analysis of Interleaved Speech-Text Language Models"](https://arxiv.org/abs/2504.02398).

We refer you to the _SlamKit_ [codebase](https://github.com/slp-rl/slamkit) to see how you can evaluate your SpeechLM with this dataset.

## sSC and tSC
We split the generation for spoken-stroycloze and topic-storycloze as detailed in [Twist](https://arxiv.org/abs/2305.13009).

## Usage

```python
from datasets import load_dataset
dataset = load_dataset("slprl/multispeaker-storycloze")
```
## Data Fields

The data has several fields:
- `id`: the file id as in the original [StoryCloze]() dataset.
- `correct_text`: the text of the correct sample.
- `correct_audio`: the synthesized audio of the correct sample.
- `incorrect_text`: the text of the incorrect sample.
- `incorrect_audio`: the synthesized audio of the incorrect sample.


## Citation
If you use this version of the dataset please cite our work:

```
@misc{maimon2025scaling,
      title={Scaling Analysis of Interleaved Speech-Text Language Models}, 
      author={Gallil Maimon and Michael Hassid and Amit Roth and Yossi Adi},
      year={2025},
      eprint={2504.02398},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2504.02398}, 
}
```