Add files using upload-large-folder tool
Browse files- README.md +5 -40
- data/In_domain_balanced_set/test-00000-of-00001.parquet +2 -2
- data/StyleInstructCaps-MetaSet/test-00000-of-00001.parquet +2 -2
- data/StyleInstructCaps/eval-00000-of-00001.parquet +2 -2
- data/StyleInstructCaps/train-00000-of-00011.parquet +2 -2
- data/StyleInstructCaps/train-00001-of-00011.parquet +2 -2
- data/StyleInstructCaps/train-00002-of-00011.parquet +2 -2
- data/StyleInstructCaps/train-00003-of-00011.parquet +2 -2
- data/StyleInstructCaps/train-00004-of-00011.parquet +2 -2
- data/StyleInstructCaps/train-00005-of-00011.parquet +2 -2
- data/StyleInstructCaps/train-00006-of-00011.parquet +2 -2
- data/StyleInstructCaps/train-00007-of-00011.parquet +2 -2
- data/StyleInstructCaps/train-00008-of-00011.parquet +2 -2
- data/StyleInstructCaps/train-00009-of-00011.parquet +2 -2
- data/StyleInstructCaps/train-00010-of-00011.parquet +2 -2
- scripts/prepare_release.py +1 -28
README.md
CHANGED
|
@@ -35,22 +35,20 @@ configs:
|
|
| 35 |
|
| 36 |
# StyleInstructCaps
|
| 37 |
|
| 38 |
-
Style-captioned speech metadata for speaking-style captioning and instruction-following evaluation
|
| 39 |
-
|
| 40 |
-
**Repository placeholder:** `<HF_ORG>/<DATASET_NAME>`
|
| 41 |
|
| 42 |
## Overview
|
| 43 |
|
| 44 |
This dataset release contains speaking-style captions, caption-generation instructions (when available), and minimal utterance metadata for speech captioning research. Each config includes only:
|
| 45 |
|
| 46 |
-
- `source`, `relative_audio_path`, `transcription`, `speakerid`
|
| 47 |
- Speaking-style caption fields (when present)
|
| 48 |
- Instruction fields for caption generation (when present)
|
| 49 |
|
| 50 |
Configs included:
|
| 51 |
|
| 52 |
- **StyleInstructCaps** — large-scale training and evaluation splits
|
| 53 |
-
- **StyleInstructCaps-MetaSet**
|
| 54 |
- **In-domain balanced set** — accent/gender-balanced in-distribution evaluation set
|
| 55 |
- **VCTK speaker prompts** — VCTK evaluation with 10 speaker-idiosyncratic style prompts
|
| 56 |
- **VCTK (unseen prompts)** — VCTK out-of-distribution prompt evaluation (32 utterances, 36 prompts per row)
|
|
@@ -58,8 +56,6 @@ Configs included:
|
|
| 58 |
|
| 59 |
This is a **metadata-only** release. Audio is **not** redistributed. Each row includes a `relative_audio_path` that should be resolved against the corresponding upstream audio corpus listed below.
|
| 60 |
|
| 61 |
-
During release preparation, local absolute `audio_path` values are removed, mixed string/list caption fields are normalized to consistent types, and `duration` is stored as `float64`.
|
| 62 |
-
|
| 63 |
## License
|
| 64 |
|
| 65 |
All resources are released under the [CC BY-NC 4.0](https://creativecommons.org/licenses/by-nc/4.0/) license.
|
|
@@ -96,7 +92,6 @@ These columns appear in a config when present in the source manifest:
|
|
| 96 |
| `relative_audio_path` | string | Relative path to the audio file within the upstream corpus |
|
| 97 |
| `transcription` | string | Transcript of the utterance |
|
| 98 |
| `speakerid` | string | Speaker identifier |
|
| 99 |
-
| `duration` | float | Utterance duration in seconds |
|
| 100 |
|
| 101 |
Style caption and instruction columns are config-specific; see below.
|
| 102 |
|
|
@@ -111,7 +106,6 @@ Style caption and instruction columns are config-specific; see below.
|
|
| 111 |
|
| 112 |
| Column | Type | Description |
|
| 113 |
| ------ | ---- | ----------- |
|
| 114 |
-
| `text_description` | list[string] | Natural-language style caption(s) |
|
| 115 |
| `speaker_idiosyncratic_style` | string | Speaker identity style caption |
|
| 116 |
| `situational_contextual_style` | string | Situational/contextual style caption |
|
| 117 |
| `expressive_emotional_style` | string | Expressive/emotional style caption |
|
|
@@ -134,7 +128,7 @@ Style caption and instruction columns are config-specific; see below.
|
|
| 134 |
|
| 135 |
| Split | Rows | Description |
|
| 136 |
| ----- | ---- | ----------- |
|
| 137 |
-
| `test` | 434 |
|
| 138 |
|
| 139 |
Schema matches **StyleInstructCaps/eval** (base columns, all style caption fields, and all six `*_instruction` columns).
|
| 140 |
|
|
@@ -144,7 +138,7 @@ Schema matches **StyleInstructCaps/eval** (base columns, all style caption field
|
|
| 144 |
| ----- | ---- | ----------- |
|
| 145 |
| `test` | 400 | Accent/gender-balanced in-distribution evaluation set |
|
| 146 |
|
| 147 |
-
**Columns:** base metadata (`source`, `relative_audio_path`, `speakerid`
|
| 148 |
|
| 149 |
### VCTK speaker prompts
|
| 150 |
|
|
@@ -189,32 +183,3 @@ Map each row's `source` and `relative_audio_path` to the corresponding upstream
|
|
| 189 |
| `ears` | [EARS Dataset](https://github.com/facebookresearch/ears_dataset) |
|
| 190 |
| `expresso` | [Expresso](https://github.com/facebookresearch/textlesslib/tree/main/examples/expresso/dataset) |
|
| 191 |
| `eval_giga` | [GigaSpeech](https://huggingface.co/datasets/speechcolab/gigaspeech) (evaluation subset) |
|
| 192 |
-
|
| 193 |
-
## Preparing and Uploading (maintainers)
|
| 194 |
-
|
| 195 |
-
From the project root:
|
| 196 |
-
|
| 197 |
-
```bash
|
| 198 |
-
python -m venv .venv
|
| 199 |
-
source .venv/bin/activate
|
| 200 |
-
pip install -r release/scripts/requirements.txt
|
| 201 |
-
|
| 202 |
-
python release/scripts/prepare_release.py
|
| 203 |
-
|
| 204 |
-
# Edit release/README.md: replace <HF_ORG>/<DATASET_NAME> and citation
|
| 205 |
-
huggingface-cli login
|
| 206 |
-
python release/scripts/upload_to_hf.py --org <HF_ORG> --name <DATASET_NAME>
|
| 207 |
-
```
|
| 208 |
-
|
| 209 |
-
## Citation
|
| 210 |
-
|
| 211 |
-
```bibtex
|
| 212 |
-
@misc{styleinstructcaps2026,
|
| 213 |
-
title = {StyleInstructCaps: Style-Captioned Speech Annotations},
|
| 214 |
-
author = {<TODO: authors>},
|
| 215 |
-
year = {2026},
|
| 216 |
-
howpublished = {\url{https://huggingface.co/datasets/<HF_ORG>/<DATASET_NAME>}}
|
| 217 |
-
}
|
| 218 |
-
```
|
| 219 |
-
|
| 220 |
-
Replace `<HF_ORG>`, `<DATASET_NAME>`, and the citation fields before publishing.
|
|
|
|
| 35 |
|
| 36 |
# StyleInstructCaps
|
| 37 |
|
| 38 |
+
Style-captioned speech metadata for speaking-style captioning and instruction-following evaluation.
|
|
|
|
|
|
|
| 39 |
|
| 40 |
## Overview
|
| 41 |
|
| 42 |
This dataset release contains speaking-style captions, caption-generation instructions (when available), and minimal utterance metadata for speech captioning research. Each config includes only:
|
| 43 |
|
| 44 |
+
- `source`, `relative_audio_path`, `transcription`, `speakerid` (when present in the source manifest)
|
| 45 |
- Speaking-style caption fields (when present)
|
| 46 |
- Instruction fields for caption generation (when present)
|
| 47 |
|
| 48 |
Configs included:
|
| 49 |
|
| 50 |
- **StyleInstructCaps** — large-scale training and evaluation splits
|
| 51 |
+
- **StyleInstructCaps-MetaSet** (434 utterances)
|
| 52 |
- **In-domain balanced set** — accent/gender-balanced in-distribution evaluation set
|
| 53 |
- **VCTK speaker prompts** — VCTK evaluation with 10 speaker-idiosyncratic style prompts
|
| 54 |
- **VCTK (unseen prompts)** — VCTK out-of-distribution prompt evaluation (32 utterances, 36 prompts per row)
|
|
|
|
| 56 |
|
| 57 |
This is a **metadata-only** release. Audio is **not** redistributed. Each row includes a `relative_audio_path` that should be resolved against the corresponding upstream audio corpus listed below.
|
| 58 |
|
|
|
|
|
|
|
| 59 |
## License
|
| 60 |
|
| 61 |
All resources are released under the [CC BY-NC 4.0](https://creativecommons.org/licenses/by-nc/4.0/) license.
|
|
|
|
| 92 |
| `relative_audio_path` | string | Relative path to the audio file within the upstream corpus |
|
| 93 |
| `transcription` | string | Transcript of the utterance |
|
| 94 |
| `speakerid` | string | Speaker identifier |
|
|
|
|
| 95 |
|
| 96 |
Style caption and instruction columns are config-specific; see below.
|
| 97 |
|
|
|
|
| 106 |
|
| 107 |
| Column | Type | Description |
|
| 108 |
| ------ | ---- | ----------- |
|
|
|
|
| 109 |
| `speaker_idiosyncratic_style` | string | Speaker identity style caption |
|
| 110 |
| `situational_contextual_style` | string | Situational/contextual style caption |
|
| 111 |
| `expressive_emotional_style` | string | Expressive/emotional style caption |
|
|
|
|
| 128 |
|
| 129 |
| Split | Rows | Description |
|
| 130 |
| ----- | ---- | ----------- |
|
| 131 |
+
| `test` | 434 | StyleInstructCaps-MetaSet evaluation subset |
|
| 132 |
|
| 133 |
Schema matches **StyleInstructCaps/eval** (base columns, all style caption fields, and all six `*_instruction` columns).
|
| 134 |
|
|
|
|
| 138 |
| ----- | ---- | ----------- |
|
| 139 |
| `test` | 400 | Accent/gender-balanced in-distribution evaluation set |
|
| 140 |
|
| 141 |
+
**Columns:** base metadata (`source`, `relative_audio_path`, `speakerid`) plus `prompt1_speaker_idiosyncratic_style_instruction` through `prompt10_speaker_idiosyncratic_style_instruction`.
|
| 142 |
|
| 143 |
### VCTK speaker prompts
|
| 144 |
|
|
|
|
| 183 |
| `ears` | [EARS Dataset](https://github.com/facebookresearch/ears_dataset) |
|
| 184 |
| `expresso` | [Expresso](https://github.com/facebookresearch/textlesslib/tree/main/examples/expresso/dataset) |
|
| 185 |
| `eval_giga` | [GigaSpeech](https://huggingface.co/datasets/speechcolab/gigaspeech) (evaluation subset) |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
data/In_domain_balanced_set/test-00000-of-00001.parquet
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:08703b00146ec58532a3f2d5cdeeffe89a4d9b96eac51d8ec9a5871d8c0e1a42
|
| 3 |
+
size 28049
|
data/StyleInstructCaps-MetaSet/test-00000-of-00001.parquet
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eb4afc31fd35b743a07a35a028119f6f834d8573a15fe39d7cfcae5d04ae37a4
|
| 3 |
+
size 316893
|
data/StyleInstructCaps/eval-00000-of-00001.parquet
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:24c100471a38e1985df6e2ebec5cb21ec9414b426232520da3754683dbc8ff04
|
| 3 |
+
size 9961717
|
data/StyleInstructCaps/train-00000-of-00011.parquet
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:45f28bd4c47fb1e66a50ff785d5aeebf19355ae27e541764564a3c348028ca96
|
| 3 |
+
size 64629185
|
data/StyleInstructCaps/train-00001-of-00011.parquet
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1d193ff3782f0f65fa580aefb8b99032fdc2de039eb8664b325dfc69ceaad768
|
| 3 |
+
size 64654933
|
data/StyleInstructCaps/train-00002-of-00011.parquet
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aeac2b6392cbdc8aa8738761713161bf04bc53f08cd79d0b651ecd5b45d2f5fe
|
| 3 |
+
size 64728994
|
data/StyleInstructCaps/train-00003-of-00011.parquet
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e88e773837860025a235f7ae42612582978f8c0e694d8b58ff5ce16fd6f07487
|
| 3 |
+
size 64649856
|
data/StyleInstructCaps/train-00004-of-00011.parquet
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1eb0bd98d7eea5f09469efa876c674f4520d861da7586e295528bfe299cdd41d
|
| 3 |
+
size 64776387
|
data/StyleInstructCaps/train-00005-of-00011.parquet
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ba95fbafb1207541d05fb88d8f8d0869501b8ad6a071773f32062046278eaa5a
|
| 3 |
+
size 64724113
|
data/StyleInstructCaps/train-00006-of-00011.parquet
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fa94e9ec14fa3f69defb4f531ce7b1081fde86082952f6ab8f93bc142a08fb36
|
| 3 |
+
size 64738026
|
data/StyleInstructCaps/train-00007-of-00011.parquet
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:825b9883168e21084de28988f5a918d9ea9506cfedcaf850045be391cd41c6f2
|
| 3 |
+
size 64759883
|
data/StyleInstructCaps/train-00008-of-00011.parquet
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:01073126644e9591f84fc4ac9d81333fe671a76c75fea8c90d54d306bf7471ef
|
| 3 |
+
size 64664058
|
data/StyleInstructCaps/train-00009-of-00011.parquet
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6e67c3bd8a01382e2e443ad775a6a87417a001c97371292539e4f45dd07a2a11
|
| 3 |
+
size 64498536
|
data/StyleInstructCaps/train-00010-of-00011.parquet
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bb90e1df162c0733690c09a533508c481028e3284d71ea08de05b44b0e3b5086
|
| 3 |
+
size 20532521
|
scripts/prepare_release.py
CHANGED
|
@@ -97,11 +97,9 @@ BASE_FIELDS = {
|
|
| 97 |
"relative_audio_path",
|
| 98 |
"transcription",
|
| 99 |
"speakerid",
|
| 100 |
-
"duration",
|
| 101 |
}
|
| 102 |
|
| 103 |
STYLE_CAPTION_FIELDS = {
|
| 104 |
-
"text_description",
|
| 105 |
"speaker_idiosyncratic_style",
|
| 106 |
"situational_contextual_style",
|
| 107 |
"expressive_emotional_style",
|
|
@@ -112,9 +110,7 @@ STYLE_CAPTION_FIELDS = {
|
|
| 112 |
"captions",
|
| 113 |
}
|
| 114 |
|
| 115 |
-
STYLE_TEXT_FIELDS = STYLE_CAPTION_FIELDS
|
| 116 |
-
|
| 117 |
-
LIST_FIELDS = {"text_description"}
|
| 118 |
|
| 119 |
|
| 120 |
def _to_text(value: Any) -> str | None:
|
|
@@ -128,22 +124,6 @@ def _to_text(value: Any) -> str | None:
|
|
| 128 |
return str(value)
|
| 129 |
|
| 130 |
|
| 131 |
-
def _to_string_list(value: Any) -> list[str]:
|
| 132 |
-
if value is None:
|
| 133 |
-
return []
|
| 134 |
-
if isinstance(value, str):
|
| 135 |
-
return [value]
|
| 136 |
-
if isinstance(value, list):
|
| 137 |
-
return [str(item) for item in value if item is not None]
|
| 138 |
-
return [str(value)]
|
| 139 |
-
|
| 140 |
-
|
| 141 |
-
def _to_float(value: Any) -> float | None:
|
| 142 |
-
if value is None:
|
| 143 |
-
return None
|
| 144 |
-
return float(value)
|
| 145 |
-
|
| 146 |
-
|
| 147 |
def is_instruction_field(key: str) -> bool:
|
| 148 |
return key.endswith("_instruction") or (
|
| 149 |
key.startswith("prompt") and "_instruction" in key
|
|
@@ -163,13 +143,6 @@ def normalize_record(record: dict[str, Any]) -> dict[str, Any]:
|
|
| 163 |
if field in cleaned:
|
| 164 |
cleaned[field] = _to_text(cleaned[field])
|
| 165 |
|
| 166 |
-
for field in LIST_FIELDS:
|
| 167 |
-
if field in cleaned:
|
| 168 |
-
cleaned[field] = _to_string_list(cleaned[field])
|
| 169 |
-
|
| 170 |
-
if "duration" in cleaned:
|
| 171 |
-
cleaned["duration"] = _to_float(cleaned["duration"])
|
| 172 |
-
|
| 173 |
return cleaned
|
| 174 |
|
| 175 |
|
|
|
|
| 97 |
"relative_audio_path",
|
| 98 |
"transcription",
|
| 99 |
"speakerid",
|
|
|
|
| 100 |
}
|
| 101 |
|
| 102 |
STYLE_CAPTION_FIELDS = {
|
|
|
|
| 103 |
"speaker_idiosyncratic_style",
|
| 104 |
"situational_contextual_style",
|
| 105 |
"expressive_emotional_style",
|
|
|
|
| 110 |
"captions",
|
| 111 |
}
|
| 112 |
|
| 113 |
+
STYLE_TEXT_FIELDS = STYLE_CAPTION_FIELDS
|
|
|
|
|
|
|
| 114 |
|
| 115 |
|
| 116 |
def _to_text(value: Any) -> str | None:
|
|
|
|
| 124 |
return str(value)
|
| 125 |
|
| 126 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 127 |
def is_instruction_field(key: str) -> bool:
|
| 128 |
return key.endswith("_instruction") or (
|
| 129 |
key.startswith("prompt") and "_instruction" in key
|
|
|
|
| 143 |
if field in cleaned:
|
| 144 |
cleaned[field] = _to_text(cleaned[field])
|
| 145 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 146 |
return cleaned
|
| 147 |
|
| 148 |
|