Datasets:
Publish verified audio data and metadata
Browse filesThis view is limited to 50 files because it contains too many changes. See raw diff
- .gitattributes +1 -0
- LICENSE +8 -0
- README.md +104 -0
- SHA256SUMS +199 -0
- SOURCE_README.md +167 -0
- SOURCE_SDD_README.md +55 -0
- audio/test-00000.parquet +3 -0
- audio/test-00001.parquet +3 -0
- audio/test-00002.parquet +3 -0
- audio/test-00003.parquet +3 -0
- audio/test-00004.parquet +3 -0
- audio/test-00005.parquet +3 -0
- audio/test-00006.parquet +3 -0
- audio/test-00007.parquet +3 -0
- audio/test-00008.parquet +3 -0
- audio/test-00009.parquet +3 -0
- audio/test-00010.parquet +3 -0
- audio/test-00011.parquet +3 -0
- audio/test-00012.parquet +3 -0
- audio/test-00013.parquet +3 -0
- audio/test-00014.parquet +3 -0
- audio/test-00015.parquet +3 -0
- audio/test-00016.parquet +3 -0
- audio/test-00017.parquet +3 -0
- audio/test-00018.parquet +3 -0
- audio/test-00019.parquet +3 -0
- audio/test-00020.parquet +3 -0
- audio/test-00021.parquet +3 -0
- audio/test-00022.parquet +3 -0
- audio/test-00023.parquet +3 -0
- audio/test-00024.parquet +3 -0
- audio/test-00025.parquet +3 -0
- audio/test-00026.parquet +3 -0
- audio/test-00027.parquet +3 -0
- audio/test-00028.parquet +3 -0
- audio/test-00029.parquet +3 -0
- audio/test-00030.parquet +3 -0
- audio/test-00031.parquet +3 -0
- audio/test-00032.parquet +3 -0
- audio/test-00033.parquet +3 -0
- audio/test-00034.parquet +3 -0
- audio/test-00035.parquet +3 -0
- audio/test-00036.parquet +3 -0
- audio/test-00037.parquet +3 -0
- audio/test-00038.parquet +3 -0
- audio/test-00039.parquet +3 -0
- audio/test-00040.parquet +3 -0
- audio/test-00041.parquet +3 -0
- audio/test-00042.parquet +3 -0
- audio/test-00043.parquet +3 -0
.gitattributes
CHANGED
|
@@ -58,3 +58,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 58 |
# Video files - compressed
|
| 59 |
*.mp4 filter=lfs diff=lfs merge=lfs -text
|
| 60 |
*.webm filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 58 |
# Video files - compressed
|
| 59 |
*.mp4 filter=lfs diff=lfs merge=lfs -text
|
| 60 |
*.webm filter=lfs diff=lfs merge=lfs -text
|
| 61 |
+
protocols/test_protocol.txt filter=lfs diff=lfs merge=lfs -text
|
LICENSE
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
cc-by-nc-4.0
|
| 2 |
+
https://creativecommons.org/licenses/by-nc/4.0/
|
| 3 |
+
|
| 4 |
+
Source: https://huggingface.co/datasets/AffectDF/AffectDF_EmotionSDD
|
| 5 |
+
|
| 6 |
+
AffectDF: Emotionally Expressive Speech Deepfake Benchmark. Original source card is retained in SOURCE_README.md.
|
| 7 |
+
|
| 8 |
+
Original source conditions apply. Audio is unchanged.
|
README.md
ADDED
|
@@ -0,0 +1,104 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-nc-4.0
|
| 3 |
+
language: [en]
|
| 4 |
+
task_categories:
|
| 5 |
+
- audio-classification
|
| 6 |
+
tags:
|
| 7 |
+
- speech-deepfake-detection
|
| 8 |
+
- emotional-speech
|
| 9 |
+
- audio
|
| 10 |
+
size_categories:
|
| 11 |
+
- 100K<n<1M
|
| 12 |
+
configs:
|
| 13 |
+
- config_name: default
|
| 14 |
+
default: true
|
| 15 |
+
data_files:
|
| 16 |
+
- split: train
|
| 17 |
+
path: audio/train-*.parquet
|
| 18 |
+
- split: validation
|
| 19 |
+
path: audio/val-*.parquet
|
| 20 |
+
- split: test
|
| 21 |
+
path: audio/test-*.parquet
|
| 22 |
+
---
|
| 23 |
+
|
| 24 |
+
# AffectDF — DeepFense data release
|
| 25 |
+
|
| 26 |
+
This repository redistributes the AffectDF audio and official protocol
|
| 27 |
+
partitions used by DeepFense. It credits the original dataset authors;
|
| 28 |
+
DeepFense is not the creator of the source recordings or spoof systems.
|
| 29 |
+
|
| 30 |
+
**Source:** https://huggingface.co/datasets/AffectDF/AffectDF_EmotionSDD
|
| 31 |
+
|
| 32 |
+
**Attribution:** AffectDF: Emotionally Expressive Speech Deepfake Benchmark. Original source card is retained in SOURCE_README.md.
|
| 33 |
+
|
| 34 |
+
**License:** [cc-by-nc-4.0](https://creativecommons.org/licenses/by-nc/4.0/). The source terms remain in force;
|
| 35 |
+
no commercial-use rights or new license are granted by this packaging.
|
| 36 |
+
All audio bytes are retained unchanged. There is no resampling, trimming,
|
| 37 |
+
normalization, denoising or lossy conversion in this release.
|
| 38 |
+
|
| 39 |
+
## Official splits
|
| 40 |
+
|
| 41 |
+
| Source split | Rows |
|
| 42 |
+
|---|---:|
|
| 43 |
+
| train | 86,999 |
|
| 44 |
+
| val | 23,330 |
|
| 45 |
+
| test | 175,468 |
|
| 46 |
+
|
| 47 |
+
Total: **285,797 recordings**. Source dev is named `val` in the original
|
| 48 |
+
SDD metadata and `validation` in the default Hugging Face dataset configuration.
|
| 49 |
+
Labels are the strings `bonafide` (genuine) and `spoof` (manipulated).
|
| 50 |
+
These are the supplied partitions; no random split was invented.
|
| 51 |
+
|
| 52 |
+
## Load portable audio
|
| 53 |
+
|
| 54 |
+
```python
|
| 55 |
+
from datasets import load_dataset
|
| 56 |
+
|
| 57 |
+
ds = load_dataset("DeepFense/AffectDF", split="train", streaming=True)
|
| 58 |
+
```
|
| 59 |
+
|
| 60 |
+
The default dataset reads `audio/*.parquet`. Each row contains an `audio`
|
| 61 |
+
feature with the **original WAV bytes**, original protocol metadata,
|
| 62 |
+
`audio_sha256`, a logical repository-relative `path`, and the original
|
| 63 |
+
absolute `sdd_path` for provenance. The audio feature loads from embedded
|
| 64 |
+
bytes; it does not require `/ds-slt` or any cluster directory. Audio decoding
|
| 65 |
+
uses the audio backend supported by your installed Datasets version.
|
| 66 |
+
|
| 67 |
+
To access the exact file bytes without decoding or resampling:
|
| 68 |
+
|
| 69 |
+
```python
|
| 70 |
+
from datasets import Audio
|
| 71 |
+
|
| 72 |
+
raw = ds.cast_column("audio", Audio(decode=False))
|
| 73 |
+
row = next(iter(raw))
|
| 74 |
+
payload = row["audio"]["bytes"]
|
| 75 |
+
```
|
| 76 |
+
|
| 77 |
+
`path` is the preserved logical WAV filename, rather than a separate loose
|
| 78 |
+
file in this repository. To recreate the local audio tree, write `payload`
|
| 79 |
+
to that relative path. `audio_sha256` verifies it against the original file.
|
| 80 |
+
The exporter verifies every embedded payload after writing each shard.
|
| 81 |
+
|
| 82 |
+
## Original SDD Parquets and protocols
|
| 83 |
+
|
| 84 |
+
- `data.parquet`: original full SDD table, copied byte-for-byte.
|
| 85 |
+
- `parquets/train.parquet`, `parquets/val.parquet`, `parquets/test.parquet`:
|
| 86 |
+
original SDD protocol files, copied byte-for-byte.
|
| 87 |
+
- `protocols/`: original source label files.
|
| 88 |
+
- `SOURCE_SDD_README.md`: local import documentation, retained as provenance.
|
| 89 |
+
- `publication.json`, `SHA256SUMS`: export checks, exact counts and checksums.
|
| 90 |
+
|
| 91 |
+
The original four metadata Parquets keep their original absolute `path`
|
| 92 |
+
and `original_path` columns. They are preserved for reproducibility and are
|
| 93 |
+
not the portable audio loader. To train existing path-based scripts on a
|
| 94 |
+
different machine, extract the embedded bytes under their logical relative
|
| 95 |
+
paths and rebase a local copy of the metadata. Do not point a new machine
|
| 96 |
+
at the original cluster paths.
|
| 97 |
+
|
| 98 |
+
## Scope
|
| 99 |
+
|
| 100 |
+
This release contains the complete imported audio and all supplied
|
| 101 |
+
train/validation/test rows. Protocol labels and source metadata are preserved.
|
| 102 |
+
Dataset-specific metadata and any source conditions are described in the
|
| 103 |
+
retained source documentation. The publisher changes storage format only;
|
| 104 |
+
it does not change the waveform bytes or claim source-dataset authorship.
|
SHA256SUMS
ADDED
|
@@ -0,0 +1,199 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
55026642310e92e4eb6eaaa2de84136f084890ad421c8d2d5a698aa5ee19715e LICENSE
|
| 2 |
+
8f1646052f4801cf1b1eff636293b28b5a2d594baf54d059d6217b2069a2a94d README.md
|
| 3 |
+
63a522476bf73a0fc0eedb6edd880b51c20314496144766d5b785816f582b9b3 SOURCE_README.md
|
| 4 |
+
ebe3d64410a6ce5284fdfbf39052741ed00321fa9a12d192ff192a3608952bdd SOURCE_SDD_README.md
|
| 5 |
+
3bce03cb4b037c943cb8925b435e489c86a76f2dcfe909310ff30314505a133b audio/test-00000.parquet
|
| 6 |
+
f7af021af79ccfe897349803f0489cce118c911b4ad6ae1add06989ddaa77d4f audio/test-00001.parquet
|
| 7 |
+
9d790a7e65d00c8e22dbb9bbd4b0121463241fcbae992e31bdb1c4171114a2a8 audio/test-00002.parquet
|
| 8 |
+
11f66b4a467533066505d6d5fe4b48712effd638c1f2e78f30b503de72d29346 audio/test-00003.parquet
|
| 9 |
+
0e3fd46aebfd938a2431a59e2209cdbca1e05d8178d845f89c2023e20fe6da76 audio/test-00004.parquet
|
| 10 |
+
b04efc3f846750b596faab943008a95f98202dff56608e5d217134c27d1350d4 audio/test-00005.parquet
|
| 11 |
+
5a91d4e93caad6c3bfa09106ab6ca013d2f46475b88d7a842d22cf63fef477af audio/test-00006.parquet
|
| 12 |
+
26d8d8156b956bb5da82a6597df201305c446582c6af143a61fbe678dfdb2871 audio/test-00007.parquet
|
| 13 |
+
767a5001fdd97f517898657fb6a9b11d4fa3bd2611b88f7e58b8c33a263f9b65 audio/test-00008.parquet
|
| 14 |
+
6571efc0583376823533c01a743bde4b9c5d897391b120682670b5623f4aa555 audio/test-00009.parquet
|
| 15 |
+
0fd4acaef9c8cfd10765c698883282cefc6cdc1a4ff764be2adc1519a83d5ad3 audio/test-00010.parquet
|
| 16 |
+
92f63102d444f921033865e4659295566fbf60ea888b595813f951e405629249 audio/test-00011.parquet
|
| 17 |
+
176d24b7882ffc7153aaca26e82e29f34009bfbb70fb6f44b70549e1b38d70dd audio/test-00012.parquet
|
| 18 |
+
b548934ccff9a333c24d9b0c21cfc3ea36ebbb94dbae5d85df1aed7435313afb audio/test-00013.parquet
|
| 19 |
+
38da447af13f00305413e3a452da4f0459dde6eb933bf0795cd9dc1dd5feb6e9 audio/test-00014.parquet
|
| 20 |
+
c3ee58c47ed4a2e16b3c26d25e1df297bad105b81c2b6c52d025a098de65d402 audio/test-00015.parquet
|
| 21 |
+
c5af005869240a6b4a4033018d99f7825587f9b41168d8be9c95efe79fe5169c audio/test-00016.parquet
|
| 22 |
+
fbdcc16867b9f66f4adbbb576d602c6c88e6e20a2ab301849312da52fe71de08 audio/test-00017.parquet
|
| 23 |
+
55dbde644869c319080bc602c19c9c8e94642010065dfc89ca1cfb390eb98eab audio/test-00018.parquet
|
| 24 |
+
122f1983dd2818732d30664b88d41029161c05e141a2ae11886b768d81ad2be0 audio/test-00019.parquet
|
| 25 |
+
5ff1e1c41b4e95fa480c3591fff8fb668792bd766fd7975a28e813efc09bfcad audio/test-00020.parquet
|
| 26 |
+
60addbf9f523824f36df2fa0147bedbd8f303f4cbd95a516ea0d57bf078a968c audio/test-00021.parquet
|
| 27 |
+
52d2f7e54ebbae8b8c174917a7a8f212469e11991b74f82735ecd80198036270 audio/test-00022.parquet
|
| 28 |
+
b1d043ac519ae6e60cca6ad7df91ebe13efc1b6862741617747acf773270e801 audio/test-00023.parquet
|
| 29 |
+
697caaf2d7abfa667a3d02a6a4149f0676e1202e374c52698edac6b49633128d audio/test-00024.parquet
|
| 30 |
+
fe57fb980fdc610fbd92ab43bf18e8d0ff418af143efb66c71a9c7537c8bb443 audio/test-00025.parquet
|
| 31 |
+
57c4e979a93e51dfe95e45f5b82e92f98ec98751bc14d8fd94475a159bd9e436 audio/test-00026.parquet
|
| 32 |
+
72dda13cbad4f9f75e4142991c7b5e7279bd2fd23b90bd44082f2ec4235f4dcc audio/test-00027.parquet
|
| 33 |
+
588bba156d5d5e52362bebccb2b2782e995d276e3a0ebb8038060ae3285c1a20 audio/test-00028.parquet
|
| 34 |
+
a17b73db9af8991dde9cdc8e0f54e9cc1e5b405297ff1e3bbfc77f30117e21d4 audio/test-00029.parquet
|
| 35 |
+
3a39f685da32242a227a91d7bf08269066e4d474f4d038586aa162cec34ffdc0 audio/test-00030.parquet
|
| 36 |
+
13cad323510ef54cadd08fc89f4451bf7338b9d1042aa70312a0de44f4f89288 audio/test-00031.parquet
|
| 37 |
+
d7bacae8780122ce817cef57202ca1583634f264003f8fb404cd19a65b6922a2 audio/test-00032.parquet
|
| 38 |
+
0b9d10de5d1b205da93c4dc14717f2c2da7ffa8b7f1dc43df333e48c3b215ed1 audio/test-00033.parquet
|
| 39 |
+
ff400c90c16a3949c6434d476d57ff5e19bd12e9f0549bccfa94ee195049e68f audio/test-00034.parquet
|
| 40 |
+
36fb9eda30584438400374de62b624a18ab6f0e1b8b15032c1c4e88ecffd5a6b audio/test-00035.parquet
|
| 41 |
+
3f8e67d6db86144115eba60c158bd97d147f3200402ad9a90e207dceb26e4c96 audio/test-00036.parquet
|
| 42 |
+
0a47cdd2221340580b30229c6efa5577620198e4e081d74a48233e6b6a2cb2b8 audio/test-00037.parquet
|
| 43 |
+
eaee1b8bdd248d8197e0d641fa1056314a4f8400a8e74a013bb9bc500e25a910 audio/test-00038.parquet
|
| 44 |
+
02f595f609314428775339e320159f0dbfb847d5f1b68c377ad1b63080c076fa audio/test-00039.parquet
|
| 45 |
+
ca612366a862bddec117685d09f2ca0001974b6d7e2ee958890502a1841d9217 audio/test-00040.parquet
|
| 46 |
+
9373af73e666eeb74e7fda199bfd00b291efbe931936d4ef4f66ffbd5e0febd0 audio/test-00041.parquet
|
| 47 |
+
ddb9e6be8bd1c4cc6ffcb084a99ac2f29e01b8ea5698bbf1f951391b543112f8 audio/test-00042.parquet
|
| 48 |
+
87980669a0ca5ecf186676db351c21447f363d57e39d30a1f11851c3167bf99b audio/test-00043.parquet
|
| 49 |
+
8b09031dad4e53ac1f412b2452aaffc4c5a49b5f2d222e6dd23b37c9ff6e6c2d audio/test-00044.parquet
|
| 50 |
+
a8b17946d89cf54392942c6dcd7dc49f94fd54627c2ec8e1e1060587227864fe audio/test-00045.parquet
|
| 51 |
+
dd3557559570474f23eeafc3ed87dbff8935e98de6182fda11e457eedd8c47d5 audio/test-00046.parquet
|
| 52 |
+
9a818d66273313bc25815e4c94ad750299bf95eef78ff9d4c6404dd3a86cc699 audio/test-00047.parquet
|
| 53 |
+
7a9069ccaec8a695671374a03a676aa410859e9934a622b35c3713cd57c09202 audio/test-00048.parquet
|
| 54 |
+
2aa0583e664db02e7df468975033e277339b01b9c088d89ab591603f85adebbe audio/test-00049.parquet
|
| 55 |
+
e2d3e1a0c127968ffc8fc8c91b71f9670df969bcfd76f4adde32ff22249fec00 audio/test-00050.parquet
|
| 56 |
+
85a4924556eb8023dfa29356d37c0ba63ee56a9bd381f438a985b6ba6bd7dbc5 audio/test-00051.parquet
|
| 57 |
+
c91df3f7cf57e45847d94a2342ca1f3d0f032ac9e7fe88ad80c84230077f7af5 audio/test-00052.parquet
|
| 58 |
+
8d15056677a4d1de12169729b21e792939715d01550d5fcc3e3d8ab52897e451 audio/test-00053.parquet
|
| 59 |
+
edaa19f4d6f18c8bbc922cfc79833719cabc1fe919fa91e7cf3fc5964e54013b audio/test-00054.parquet
|
| 60 |
+
fcb10110556b1593bb3ae31cf2f15de86a9d7a2f023b7412d88408deb36f1957 audio/test-00055.parquet
|
| 61 |
+
e6ba46e5a45f5c7228e27d58268e2ce863c4d0f1c6550f1b8f738a82470c69be audio/test-00056.parquet
|
| 62 |
+
cc90e6476f77db0092373b4ca955f0713ae5c32bf031fac9588b50a8b342823e audio/test-00057.parquet
|
| 63 |
+
07f90c9c636dcd56c7d3a521fde33bb9258947b6d58cdc4407495809858ac137 audio/test-00058.parquet
|
| 64 |
+
91fd6006e7482593f033ad368017045ba4d45f4e66bc18617d90d0ee7cdc52d3 audio/test-00059.parquet
|
| 65 |
+
29877a9b9dbbe2fe412e209dcd568b0be6fecfe61516cdbd441ddd4fa28f1e3d audio/test-00060.parquet
|
| 66 |
+
120ec35417b22829ec9e0776219fa1136c7927a297e33e4370fbd9434e63a25c audio/test-00061.parquet
|
| 67 |
+
c479181e8386f1343a61e213be20e6ddefc7c6cccd4ee1329344e845a496ef19 audio/test-00062.parquet
|
| 68 |
+
8de1753c4ce35a8e264ed41650894c0d44483c9737fb056bd387b1299ee4ff62 audio/test-00063.parquet
|
| 69 |
+
2fec8251e1919e323a93398a3cf41f903c97a263746139c67f076b5c2a3974ba audio/test-00064.parquet
|
| 70 |
+
deaebe1d93653d4d88af710627677964d03dd0d6d1e40f97e44039aa8e92a40e audio/test-00065.parquet
|
| 71 |
+
e5ba1b467a52223b7250a0a124bbae11e6aefe9ee93e8e6fb5bb1d80fd1e49d4 audio/test-00066.parquet
|
| 72 |
+
6e7ae7e8b093492775970a203ff18b3e16cf3e2f2442071548c0b41d55eb4a3c audio/test-00067.parquet
|
| 73 |
+
b54c4115fa3fbad54b3bb6734c8fd216e65f9b1170239c061c84a6afb1570106 audio/test-00068.parquet
|
| 74 |
+
ef633b8691db7c619edf8eef3d58662e2bc393875b677bcdc9250b8542fb2179 audio/test-00069.parquet
|
| 75 |
+
68d401745dbeb85e166a168661aa85e982a4b4fdca64c989518c7bdd6cb1bc4f audio/test-00070.parquet
|
| 76 |
+
3ef238e7353d8b745b6155899c06d66a26bc26e029a83e121ceeac6eef4dcb9c audio/test-00071.parquet
|
| 77 |
+
1b10d3a647c8e24c42d992a8818f2f19e433e0458e1e6f949efe22099555a467 audio/test-00072.parquet
|
| 78 |
+
3cc939e6a785482d37e4b70f78c7dc795748d968f748f6c36204f7a8138f6d25 audio/test-00073.parquet
|
| 79 |
+
b984df89108dbe341480d1fe010d3efaa73f3ce54ca459a207fa5070d5464bad audio/test-00074.parquet
|
| 80 |
+
3aa0124a36fb0f69ee419e0965c0201022dcd71aaa987b4d33be66e84b01893e audio/test-00075.parquet
|
| 81 |
+
46e91661f89d4c3c7296fa0a8b88553b71289a134168da5b6b52fcaaef1ef02f audio/test-00076.parquet
|
| 82 |
+
661c6d6f6875b21001e69333b4064b2f29d8f5e2acd070622b673ebff43ffc21 audio/test-00077.parquet
|
| 83 |
+
649bea1f573a7a6fa5d819ebcdadae9946e5fcef8b39c1f3af94a8976158a833 audio/test-00078.parquet
|
| 84 |
+
4e33d6a992b998f150f95fa11e2520a07a91168b55ae89852e2ef15636f08aab audio/test-00079.parquet
|
| 85 |
+
0148f8f41350d6cc81ab52b7802aebc74f0d38387a6ac78c1de19abfb41cd509 audio/test-00080.parquet
|
| 86 |
+
adc646be1d236b1abfe600547f7181bd54a9950a785594ab5d2112074579835c audio/test-00081.parquet
|
| 87 |
+
6c4aa79cc6843c20381b2686f74a6fc8768878f2ff3a06ee966d154b6aeb5a5d audio/test-00082.parquet
|
| 88 |
+
05df7c7e59d9db1a54c0463ab8a9c1cacbc54ed1de2f70f53699797cd3eeadc1 audio/test-00083.parquet
|
| 89 |
+
73f9dc9dac741976736c51a39083df55c3d85ac8f0ce4f08a256e0ce436e194e audio/test-00084.parquet
|
| 90 |
+
e6232f54d3320c57b923c8c403331409e357583cded14d3ab4a25e9e36635f0c audio/test-00085.parquet
|
| 91 |
+
f46339daee7aa3d82257760dc4c298d6d0fc690bef83c7bd9620923d36faf048 audio/test-00086.parquet
|
| 92 |
+
afe5365e274588d4e6d8cc4da41333492784343f53a7e8980a86e9127fe0cb6e audio/test-00087.parquet
|
| 93 |
+
691f5ae0b6444d16fc5afa2c7061624684547fdd2f58c2367a5557a2a81dec10 audio/test-00088.parquet
|
| 94 |
+
052ca8733c00b37dcb5074d52173b20d83ccd091620f2a94a865eac2bb2b2500 audio/test-00089.parquet
|
| 95 |
+
1e71ae73d963719dd45a8b738f2b78178256da11f26adbe2b0919ca125891c8c audio/test-00090.parquet
|
| 96 |
+
ecdae3092bb26b3104e2c8f662812f46dc8511f783bf3e8bc1814d25356373d3 audio/test-00091.parquet
|
| 97 |
+
16184d2b52a46bee2de0d15dc3b718091eaf01174dff1cd90cec76dc5c590292 audio/test-00092.parquet
|
| 98 |
+
9257818bbd6106987b79c0e672e8226606a94f627e1d4fc262da76dba1474bef audio/test-00093.parquet
|
| 99 |
+
88c817dd601b2a648e211dcd53f8b4489d6f77a93beb119401d1fa4243f7cfce audio/test-00094.parquet
|
| 100 |
+
0822121d30e8d70a0d4fa8f1960b64809e0348267bee4a0ff31238324fb1d7c4 audio/test-00095.parquet
|
| 101 |
+
85a38efe2b2f58308bbc305eef02588c9ed1c5993a1720bca6354278eda77a5c audio/test-00096.parquet
|
| 102 |
+
ab204eba7214e30b7b9b935a63e51acaa3e33fdde6b1f63bf928bde59a9f123c audio/test-00097.parquet
|
| 103 |
+
cddc1d19d1522aef4b60fd31a95b0c357c208182b03996fb598292a008c680ff audio/test-00098.parquet
|
| 104 |
+
7ab7521a10259bb7f88f373e4ad3781c0ce6ff51f43b845c50b4aab781370f79 audio/test-00099.parquet
|
| 105 |
+
56d4c1e94adbbdfeff1b0975d1a2aa151d8873af9ef5d452ee14b7ae84531844 audio/test-00100.parquet
|
| 106 |
+
e611d771a025b10cef4840fdfd5bd04117502bb33e0b2cb07704158495ef84d5 audio/test-00101.parquet
|
| 107 |
+
bab19695e9014cbf198ec2b53de01a7de55a1b62e8aa8b93042282f55fa44922 audio/test-00102.parquet
|
| 108 |
+
d9c360a0dd2eadb85a3ccc1def09325210d8e36bd461996e81875a33bfd582d0 audio/test-00103.parquet
|
| 109 |
+
81d127f09be305e30e78d5affa40a434c58ec09b8b29d2a2ec1c3d78620b330a audio/test-00104.parquet
|
| 110 |
+
7f407dc0b6eb2026162ba32f822dcf2f19747c172774955ca8b59f258bc4eeb7 audio/test-00105.parquet
|
| 111 |
+
2b59650209b01510657da3303a12e6a3d7b49d02774b0137b5d0b6547c4a44ae audio/test-00106.parquet
|
| 112 |
+
8d24d1b7afbe13dd2ea18c6d38785d0529b3cad27547ec2b5ceb6dc91496e49a audio/test-00107.parquet
|
| 113 |
+
2958b6fbbe507a1ad4da690a1bfc754712d03f19cd4a34f9f776c73d78741e5c audio/test-00108.parquet
|
| 114 |
+
cc93cd370849276446632dd16bed4cb84d70c8b9b5f239b9914c384c6699d2ec audio/test-00109.parquet
|
| 115 |
+
876e256e8a1f77052febb2ca36bd484ac245375dfedc60a8a80e0db2e75a5ddf audio/test-00110.parquet
|
| 116 |
+
e1a2f0681ebcc8d138ce0beca530980c20289c6b9b88f526d66f922529d5b416 audio/test-00111.parquet
|
| 117 |
+
92f1d6f20827fad2c7499ad5302f7d6c9d21b8ed158865a2e22296a61bdab749 audio/test-00112.parquet
|
| 118 |
+
e52b0a95a1bb5a23af86d9d7cd003ebdf65966bdc7ad842d759d07b320b54d68 audio/test-00113.parquet
|
| 119 |
+
efcb62a5339aae8877bb52d254a0be15ca2c017ae3cf2051f282768722a5aa5d audio/test-00114.parquet
|
| 120 |
+
39879fcc0e424ed3c86eabc367595d80ac524e4ed11746a84f5f2674b89401a5 audio/test-00115.parquet
|
| 121 |
+
e846144f5472dec64bacd12160d57de7d83a38acbe7a740aa57aa20cae2083c1 audio/test-00116.parquet
|
| 122 |
+
8490ce6eacf62ded9496f25d85e6a0d67782e5eef031fd2e10d51ef5371a234a audio/test-00117.parquet
|
| 123 |
+
9cd1fc65f3c5df2a9e5a80a927aa96863f9c7a47833bd1fbfac7263917558b20 audio/test-00118.parquet
|
| 124 |
+
e1ae219b94a1e9b56862c62136f37059c9e7fe13f19b1ddfc1150c3a406cb306 audio/test-00119.parquet
|
| 125 |
+
09c6babba48cf4ddec649db23b4edefc58c177c4c930d400b2f7af8de6318838 audio/test-00120.parquet
|
| 126 |
+
6b7caafb00c7e5863a99639a824dd0f72b16768d103ee96655b11086e4688fca audio/test-00121.parquet
|
| 127 |
+
4b5de48c4d3367ede98da1d13c9ddbdc5d791bd0967e61369e3f659098b6392b audio/test-00122.parquet
|
| 128 |
+
b9329cce1d60f20065d79f7e192984551efd79d5f21896a60f09eac51833b952 audio/test-00123.parquet
|
| 129 |
+
8bcbb023bba3b7c681ec0f055d9159a1be8b1e6890a5c5e0cd74926d4c8b516c audio/test-00124.parquet
|
| 130 |
+
054420766f500533b799357c26a9480089d8e7a5c2b64b84958aea33897a1d3a audio/test-00125.parquet
|
| 131 |
+
f33eda0c7766793f18d8f81f02cb308c91a2e19629f2de58a327c0d95b0325d4 audio/test-00126.parquet
|
| 132 |
+
529ed1d46958aac294a456dc7aeb37c4c0f1fef278f363b3343427b4a3788501 audio/test-00127.parquet
|
| 133 |
+
4b3dde52c7c615ffa4a85d684629f596a13468bd86e2cec37ad4e15b4b23565b audio/train-00000.parquet
|
| 134 |
+
661eaf3c323f1d37ff7894c8810e822b431d54897985932a2a9aaa557da60d14 audio/train-00001.parquet
|
| 135 |
+
cd4fa6cc7724bc460f0cb48b0e1e490904a0e664777bc3020e1a201624a4ce9e audio/train-00002.parquet
|
| 136 |
+
af15a77f71c017e7442575835f85b945f39c393b8272a2be383bf2cf19b73636 audio/train-00003.parquet
|
| 137 |
+
0e360dd75dd15e4cb494ede4dd37d322daa4305c1c61cf9b86479c640fd5808b audio/train-00004.parquet
|
| 138 |
+
1033a48d6bbc00231380f7e4f1d312fa2ec1e7047ac6edf23ee71f8efec3dff3 audio/train-00005.parquet
|
| 139 |
+
eba23170dc1821b39b145ecfb22203c4f3499668075b52a41986fd96b63e98ab audio/train-00006.parquet
|
| 140 |
+
e46bfbb85d5e12c4db9eedf69f5c371a28691d79fbc5a679d0b9df024b9cce4d audio/train-00007.parquet
|
| 141 |
+
b6c8a29a8efe57566abed77d217864e20c6002fe1e7651a9ebf3a6667dff56d2 audio/train-00008.parquet
|
| 142 |
+
aec6b86a835ced09e14a73d43623c0238e294f4d2a64d9db0b260cf323c7eb49 audio/train-00009.parquet
|
| 143 |
+
a2dbd589e4949878a1362f94ee0c1c325bf8945c33bc56fa8bdfae8d2c1c0fc7 audio/train-00010.parquet
|
| 144 |
+
fb29ff01c1b70021e2d692f17f34a19ca8d12ba583ca30a21e5f0129e18a0da4 audio/train-00011.parquet
|
| 145 |
+
10398ba5df231b994cb6e537d633d3ea874a972a6d0016f4b1a9bb2d4dd9cec3 audio/train-00012.parquet
|
| 146 |
+
884702a96f805999ef379e5751b549fb2c888843628e64bfdf810739f2b3f08a audio/train-00013.parquet
|
| 147 |
+
8772119192575e82335a65e3521ac8bd003611a1b2ca36c9912f7a72cf0a8fa6 audio/train-00014.parquet
|
| 148 |
+
e7cafa07196d42120fc982b8a590aab97883b78872f4644246f2fe318915c0ec audio/train-00015.parquet
|
| 149 |
+
4f91fb6ae0156597106a4df8bd5cf9a09103aa0c835f085057daa38a2eed6e35 audio/train-00016.parquet
|
| 150 |
+
27e140bb73bdb518beae5adacf40286e5fe2e0a12e2af46fa01c8ee84f9d3604 audio/train-00017.parquet
|
| 151 |
+
90a2dc0fd9b778e293df6a95126c0c2050ce898d1de11458afa48024de8b534a audio/train-00018.parquet
|
| 152 |
+
566a077b296a2643b588283b101c9759aeeb619e89f6b3ff0ceee1eeacde792d audio/train-00019.parquet
|
| 153 |
+
803d52486311e713669156e1828ffad7f8c88a7db8a594277b496935e10d4690 audio/train-00020.parquet
|
| 154 |
+
642490a93a9da1bbbcce83b322744ce0378be9a99427422cfe9b27d404271755 audio/train-00021.parquet
|
| 155 |
+
e71e907ab08203851c8477345476d246ca0ead101a9c329c017be52544ae9720 audio/train-00022.parquet
|
| 156 |
+
0b9c513464fb05454f10dde14283df45c3671bb286ea22d3523883f2abafb8c2 audio/train-00023.parquet
|
| 157 |
+
698261c16b9b497f1a4af272e22a31179ede758526f1205e6aaa00cc8d7bd5bb audio/train-00024.parquet
|
| 158 |
+
603e7e8bd2e90ce6521799c2fe319f3b94c5f12321fe3d136511a372c35ad93f audio/train-00025.parquet
|
| 159 |
+
4b232ed376bcd957b46334066f168e38deec3eee6b1ec83440578dbb4649de40 audio/train-00026.parquet
|
| 160 |
+
016bb0adc013694cf4b0c6deaeffabe9907003a85e092496f396b93f0c9bc01d audio/train-00027.parquet
|
| 161 |
+
87a318bd609c5a16e8aa0597237c7fd06b7be4e4fcec9d61dfb064d382a93bd5 audio/train-00028.parquet
|
| 162 |
+
7a115184191a09216e5a4ee60a3fd395791f6481035aa01593e9e1c1b98acd6d audio/train-00029.parquet
|
| 163 |
+
a187ae88ea8cd6fcf7a15c549275a2389fc2e7ca917e0da8deba6490f5523b88 audio/train-00030.parquet
|
| 164 |
+
8644a0bdb54bf790b40b1e54ad3a0f20e825cc4eda432f9b80c2a0c505d2c258 audio/train-00031.parquet
|
| 165 |
+
f0808974f4d08ff19c539657fec2b6815e900cb56fb34b61545fc479986bd4ac audio/train-00032.parquet
|
| 166 |
+
703841913ea19b10d99e2cc4b6fd61e71c98db7de55cbbdac850984a4f38ee30 audio/train-00033.parquet
|
| 167 |
+
b4a32a773f2021618f9bf8ddb15bbb81e3659fc5318c3c25efeec2064e8e1778 audio/train-00034.parquet
|
| 168 |
+
d6b32d7cb59fb4dcfcf60031bf331c285a18f69f1cc303fd0ec766ee68deec72 audio/train-00035.parquet
|
| 169 |
+
c35a5656701ddfd24b6ee2c8e5bb79219e77da573956915ea70a5950ebef72d4 audio/train-00036.parquet
|
| 170 |
+
acd01fe018c1b34838210d33c847f3636f0daa48eb0cfb6901aadd1deac8c484 audio/train-00037.parquet
|
| 171 |
+
b34f2900e524a36adb1773083b59d94c6eaea565f900d9cf45bfaa76122b210f audio/train-00038.parquet
|
| 172 |
+
82ccf653b2d4031e4b2fc46afe7e0cf26d397ad0cb3f3bc1f2aaedd2092ba958 audio/train-00039.parquet
|
| 173 |
+
d26037f528dd2ccb010c8223ac11644071318dcafc4231308fa8b0b156535672 audio/train-00040.parquet
|
| 174 |
+
75d2e87a07547ba2506deddd0e2bc35206fa27c6dd862181bade8b411db83fd3 audio/train-00041.parquet
|
| 175 |
+
5b7ec98e70285d82443709f981c1c4d49a28d8966aa9d72ff05d580b053c225f audio/train-00042.parquet
|
| 176 |
+
1024dffd9579ebd84bfe1c1003209163def4f76529e0ca047c9ab44b9933fbd7 audio/train-00043.parquet
|
| 177 |
+
7d9d0dcf6a0441b52d29f893ad7d7fa76a1142c815ae8eb60d0000aa91aeb555 audio/train-00044.parquet
|
| 178 |
+
0c13fde7b23482e8c008684e021fabf86722cc63e7c4423fa94da63f32e92f88 audio/train-00045.parquet
|
| 179 |
+
a15a328a4c67fcdd5db8b47037c7afcf980c30667e2bc08c0b2a6be69a3b44be audio/val-00000.parquet
|
| 180 |
+
055e2a8221a8e992b20360039a0da62ec2cf5aa62b4fc6aeb075d923761f97f7 audio/val-00001.parquet
|
| 181 |
+
e00beb67cf1a49c3730e7d49960626908b3f1edb82abb834e47c6b89d2a44af7 audio/val-00002.parquet
|
| 182 |
+
b013161d8f68b9378feb8712bba41fd6dd1f7b0d00a415cd3efedb1e86a0ad65 audio/val-00003.parquet
|
| 183 |
+
2a1c727e5697a90a795942388fce0642b5f0b6f1134f2a57dfd2161bb7215c49 audio/val-00004.parquet
|
| 184 |
+
9a2edc921ce67df9a4b23d6c0f173364d72006e9e72ff19cd22d18a62bf52fb8 audio/val-00005.parquet
|
| 185 |
+
5080f5d1581ee007f5360acc105c828e0453e0d23bb2032947ca3d77bf58e519 audio/val-00006.parquet
|
| 186 |
+
69655cb77a8427d552e972b2a68883bc2144975700ae0b821d329596d1056d7c audio/val-00007.parquet
|
| 187 |
+
53d23aa8cfb9f8bb32a292e75989aa93a85812cb91c9221c1253626cae061510 audio/val-00008.parquet
|
| 188 |
+
6240dbf768632be079e1b8b7d5bd66710976f04d95c1ef1423c50e5889d6b449 audio/val-00009.parquet
|
| 189 |
+
8224a636ba2d9bc7d4b0111f2cf83cfb8d66ec1d07e2eb2babbb76a4f0c9b05f audio/val-00010.parquet
|
| 190 |
+
e000566400c8a65b972d07603a7c73a6cf5f42112b7d7c93841b69b09b8d9e71 audio/val-00011.parquet
|
| 191 |
+
cea350ca673ca889df3fc04264033ed5f344088661dbef298189107ab86c57f1 data.parquet
|
| 192 |
+
8815489536fcf369d7f06d3dc4e67109964f2e7deb54a7c9536432761ae94273 parquets/test.parquet
|
| 193 |
+
79c3757a1995747f950bae557fddd2e27a44c1bdd673b593fe5271835fe02b9a parquets/train.parquet
|
| 194 |
+
05f8e90c32e27cdbda8ce3065bb463e7cda232ebd7c5f3a4a48760dbae0d6167 parquets/val.parquet
|
| 195 |
+
5de0b834bfcff1e2aa652dcc4b7c8b212874c8842341b8c9035d15cc5b3bf71b protocols/dev_protocol.txt
|
| 196 |
+
69d3570cf77aa980cab38a17a22434f3276427cdf8dd430016fb836db260b575 protocols/test_protocol.txt
|
| 197 |
+
fb7ed4c4a8efe4ef022e45fe3abbfe65e323843aef87ff4f749345c5f27422cb protocols/train_protocol.txt
|
| 198 |
+
f69ba00e43083549183908f13e16663e72f19e2eef9bdb6b4e99a1eb3dd68f1d publication.json
|
| 199 |
+
4d29cd17963f83d2453e0c28ab944c744886e479dc7cb18e7090f575ae3f8db0 source-protocol-hashes.json
|
SOURCE_README.md
ADDED
|
@@ -0,0 +1,167 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-nc-4.0
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
task_categories:
|
| 6 |
+
- audio-classification
|
| 7 |
+
tags:
|
| 8 |
+
- speech-deepfake-detection
|
| 9 |
+
- audio-deepfake-detection
|
| 10 |
+
- emotional-speech
|
| 11 |
+
- spoofing-detection
|
| 12 |
+
- speech-synthesis
|
| 13 |
+
pretty_name: AffectDF EmotionSDD
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
# AffectDF: Emotionally Expressive Speech Deepfake Benchmark
|
| 17 |
+
|
| 18 |
+
## Overview
|
| 19 |
+
|
| 20 |
+
AffectDF is a large-scale benchmark for speech deepfake detection under emotionally expressive spoofing conditions. The dataset is designed to evaluate whether current speech deepfake detection (SDD) systems can generalize beyond conventional neutral-speech benchmarks to modern emotional and expressive speech attacks.
|
| 21 |
+
|
| 22 |
+
AffectDF contains approximately 260 hours of audio generated using 21 spoofing attacks across five emotional states. The dataset includes both acted emotional speech and spontaneous emotional speech, enabling analysis of robustness across different emotional speaking styles.
|
| 23 |
+
|
| 24 |
+
The dataset is organized into three official splits:
|
| 25 |
+
|
| 26 |
+
* Train/
|
| 27 |
+
* Dev/
|
| 28 |
+
* Test/
|
| 29 |
+
|
| 30 |
+
Protocol files are provided for reproducible evaluation.
|
| 31 |
+
|
| 32 |
+
---
|
| 33 |
+
|
| 34 |
+
## Source Corpora
|
| 35 |
+
|
| 36 |
+
AffectDF is constructed using bona fide emotional speech from two source corpora:
|
| 37 |
+
|
| 38 |
+
* ESD (Emotional Speech Dataset): acted emotional speech
|
| 39 |
+
* MSP-Podcast: spontaneous emotional speech
|
| 40 |
+
|
| 41 |
+
The use of both corpora enables evaluation across controlled acted emotional speech and naturalistic spontaneous emotional speech.
|
| 42 |
+
|
| 43 |
+
---
|
| 44 |
+
|
| 45 |
+
## Emotion Categories
|
| 46 |
+
|
| 47 |
+
AffectDF covers five emotional states:
|
| 48 |
+
|
| 49 |
+
* Angry
|
| 50 |
+
* Happy
|
| 51 |
+
* Neutral
|
| 52 |
+
* Sad
|
| 53 |
+
* Surprise
|
| 54 |
+
|
| 55 |
+
---
|
| 56 |
+
|
| 57 |
+
## Dataset Design
|
| 58 |
+
|
| 59 |
+
AffectDF contains both bona fide and synthetic emotional speech samples.
|
| 60 |
+
|
| 61 |
+
* Bona fide samples are taken from ESD and MSP-Podcast.
|
| 62 |
+
* Synthetic samples are generated using diverse spoofing systems spanning TTS, VC, EVC, and LALM-based attacks.
|
| 63 |
+
* The dataset supports analysis across emotion labels, attack types, generation models, speakers, and acted versus spontaneous emotional speech.
|
| 64 |
+
* Many samples follow a parallel or controlled design, where synthetic samples can be aligned with corresponding source utterances using audio IDs and protocol metadata.
|
| 65 |
+
|
| 66 |
+
---
|
| 67 |
+
|
| 68 |
+
## Spoofing Attack Types
|
| 69 |
+
|
| 70 |
+
AffectDF includes attacks from multiple generation paradigms:
|
| 71 |
+
|
| 72 |
+
* TTS: text-to-speech attacks
|
| 73 |
+
* VC: voice conversion attacks
|
| 74 |
+
* EVC: emotional voice conversion attacks
|
| 75 |
+
* VC+EVC: combined voice and emotional voice conversion attacks
|
| 76 |
+
* LALM-EVC: large audio-language model based emotional voice conversion attacks
|
| 77 |
+
|
| 78 |
+
The benchmark includes both conventional generation systems and recent large audio-language model based attacks.
|
| 79 |
+
|
| 80 |
+
---
|
| 81 |
+
|
| 82 |
+
## Directory Structure
|
| 83 |
+
|
| 84 |
+
The dataset is organized as follows:
|
| 85 |
+
|
| 86 |
+
```text
|
| 87 |
+
AffectDF/
|
| 88 |
+
├── Train/
|
| 89 |
+
├── Dev/
|
| 90 |
+
├── Test/
|
| 91 |
+
└── Protocols/
|
| 92 |
+
├── train_protocol.txt
|
| 93 |
+
├── dev_protocol.txt
|
| 94 |
+
└── test_protocol.txt
|
| 95 |
+
```
|
| 96 |
+
|
| 97 |
+
Example audio filenames:
|
| 98 |
+
|
| 99 |
+
```text
|
| 100 |
+
R_0015_000172.wav
|
| 101 |
+
CV_0016_001189.wav
|
| 102 |
+
A2S_0015_0015_000654_Angry_to_Sad.wav
|
| 103 |
+
MCPM_A2N_0017_0017_000613_Angry_to_Neutral.wav
|
| 104 |
+
StA_N2A_0015_000197.wav
|
| 105 |
+
```
|
| 106 |
+
|
| 107 |
+
---
|
| 108 |
+
|
| 109 |
+
## Protocol Files
|
| 110 |
+
|
| 111 |
+
The `Protocols/` directory contains split-specific metadata files:
|
| 112 |
+
|
| 113 |
+
```text
|
| 114 |
+
Protocols/train_protocol.txt
|
| 115 |
+
Protocols/dev_protocol.txt
|
| 116 |
+
Protocols/test_protocol.txt
|
| 117 |
+
```
|
| 118 |
+
|
| 119 |
+
Each protocol entry contains metadata needed for evaluation and analysis.
|
| 120 |
+
|
| 121 |
+
Example entry:
|
| 122 |
+
|
| 123 |
+
```text
|
| 124 |
+
0011 A2A_0011_0011_000351_Angry_to_Angry A01 Angry EVC Qwen2.5-Omni spoof
|
| 125 |
+
```
|
| 126 |
+
|
| 127 |
+
The fields include:
|
| 128 |
+
|
| 129 |
+
```text
|
| 130 |
+
speaker_id audio_id attack_id emotion generation_type generation_model label
|
| 131 |
+
```
|
| 132 |
+
|
| 133 |
+
Where:
|
| 134 |
+
|
| 135 |
+
* speaker_id: speaker identifier
|
| 136 |
+
* audio_id: utterance or file identifier
|
| 137 |
+
* attack_id: attack system identifier
|
| 138 |
+
* emotion: emotion label
|
| 139 |
+
* generation_type: attack category, such as TTS, VC, EVC, or LALM-EVC
|
| 140 |
+
* generation_model: model used to generate the sample
|
| 141 |
+
* label: real or spoof
|
| 142 |
+
|
| 143 |
+
---
|
| 144 |
+
|
| 145 |
+
## Bona Fide and Synthetic Pairing
|
| 146 |
+
|
| 147 |
+
AffectDF supports controlled comparison between bona fide and synthetic speech. Synthetic samples can be linked to corresponding or related bona fide samples through shared audio ID components in the filenames and protocol entries.
|
| 148 |
+
|
| 149 |
+
This design enables:
|
| 150 |
+
|
| 151 |
+
* real-versus-spoof comparison
|
| 152 |
+
* emotion-wise analysis
|
| 153 |
+
* attack-wise analysis
|
| 154 |
+
* acted versus spontaneous comparison
|
| 155 |
+
|
| 156 |
+
---
|
| 157 |
+
|
| 158 |
+
## Ethical Considerations
|
| 159 |
+
|
| 160 |
+
In this work, we use publicly available datasets and models to generate synthetic speech for the purpose of improving speech deepfake detection. We do not release any additional speaker information beyond what is already available in the original source datasets, and we do not synthesize samples containing harmful or sensitive content. The generated data and analyses are intended solely to support research on robust detection, security, and responsible benchmarking of speech deepfakes. We do not intend for AffectDF or any generation pipeline described in this work to be used for impersonation, deception, or other malicious applications.
|
| 161 |
+
|
| 162 |
+
---
|
| 163 |
+
|
| 164 |
+
|
| 165 |
+
## Data License and Intended Use
|
| 166 |
+
|
| 167 |
+
Data License and Intended Use. AffectDF is released for research and non-commercial use only under a CC BY-NC 4.0 license. The release includes generated audio and protocol files, subject to the licenses and terms of the original source corpora and generation models. The dataset is intended only for speech deepfake detection research and must not be used for impersonation, deception, biometric spoofing, or other harmful applications.
|
SOURCE_SDD_README.md
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# AffectDF
|
| 2 |
+
|
| 3 |
+
285,797 labelled recordings imported from `/ds-slt/audio/AffectDF`.
|
| 4 |
+
The provided protocols define the partitions; source dev maps to validation.
|
| 5 |
+
|
| 6 |
+
| Table | Rows | Bonafide | Spoof |
|
| 7 |
+
|---|---:|---:|---:|
|
| 8 |
+
| `data.parquet` | 285,797 | 19,108 | 266,689 |
|
| 9 |
+
| `parquets/train.parquet` | 86,999 | 6,000 | 80,999 |
|
| 10 |
+
| `parquets/val.parquet` | 23,330 | 3,000 | 20,330 |
|
| 11 |
+
| `parquets/test.parquet` | 175,468 | 10,108 | 165,360 |
|
| 12 |
+
|
| 13 |
+
`data.parquet` retains all supplied protocol rows and metadata. Its three
|
| 14 |
+
partitions are in `parquets/`. Every audio `path` points inside this dataset's
|
| 15 |
+
`wav/` tree; `original_path` records the original location for provenance.
|
| 16 |
+
Audio files are independent local copies in SDD.
|
| 17 |
+
|
| 18 |
+
## Metadata
|
| 19 |
+
|
| 20 |
+
`ID` is unique within the dataset and includes the language and source partition.
|
| 21 |
+
`original_ID` preserves the source waveform stem. `label` uses bonafide/spoof;
|
| 22 |
+
`source_label` preserves the original label. `dataset_name`, `subset`, `language`,
|
| 23 |
+
and `split` retain the dataset, original subset, language, and canonical partition.
|
| 24 |
+
|
| 25 |
+
AffectDF also preserves `speaker_id`, `attack_id`, `emotion`, `generation_type`,
|
| 26 |
+
and `generation_model` exactly from its seven-field protocols. Source label
|
| 27 |
+
`real` maps to `bonafide`. Audio lives in `wav/Train/`, `wav/Dev/`, and `wav/Test/`.
|
| 28 |
+
Original protocols are in `protocols/`. The source documentation and stated
|
| 29 |
+
CC BY-NC 4.0 license are retained in [SOURCE_README.md](SOURCE_README.md).
|
| 30 |
+
|
| 31 |
+
Some original local audio was owner-only and unreadable. The audio was
|
| 32 |
+
recovered from the [public release](https://huggingface.co/datasets/AffectDF/AffectDF_EmotionSDD) at revision
|
| 33 |
+
`529ed470208e52e59c911a531cf803f8e42c8755`. All three public protocols match the local protocol
|
| 34 |
+
rows. Every recovered filename and size was checked, and 30 readable source
|
| 35 |
+
recordings were compared byte-for-byte. `original_path` remains a provenance
|
| 36 |
+
field; training uses the accessible SDD `path`. Full evidence is in
|
| 37 |
+
`/ds-slt/audio/sdd-research-baseline/reports/affectdf-public-recovery.json`.
|
| 38 |
+
|
| 39 |
+
## Verification and training
|
| 40 |
+
|
| 41 |
+
Every protocol row was checked against the copied audio inventory. Class-based
|
| 42 |
+
full-waveform samples were decoded for each source subset. A subsequent audit
|
| 43 |
+
checked every parquet row, membership of each partition in the full table,
|
| 44 |
+
canonical split names, audio path prefixes, split path overlap, and sampled
|
| 45 |
+
audio reads. These checks are recorded in
|
| 46 |
+
`/ds-slt/audio/sdd-research-baseline/reports/emotional-datasets-audit.json`.
|
| 47 |
+
They do not constitute an exhaustive decode of every recording.
|
| 48 |
+
|
| 49 |
+
`import.json` records source protocol SHA-256 hashes, exact counts, copy details,
|
| 50 |
+
source-ID overlap checks, and any audio without protocol metadata.
|
| 51 |
+
|
| 52 |
+
Separate AASIST and MLP experiments use this dataset's own train/validation
|
| 53 |
+
partitions. Test data stays held out. Configs and commands are in
|
| 54 |
+
`/ds-slt/audio/yelkheir/DeepFense/deepfense/config/experiments/EmotionalSDD/README.md`.
|
| 55 |
+
The handover [runbook](../../sdd-research-baseline/README.md) records their Slurm jobs.
|
audio/test-00000.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3bce03cb4b037c943cb8925b435e489c86a76f2dcfe909310ff30314505a133b
|
| 3 |
+
size 230653254
|
audio/test-00001.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f7af021af79ccfe897349803f0489cce118c911b4ad6ae1add06989ddaa77d4f
|
| 3 |
+
size 216913539
|
audio/test-00002.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9d790a7e65d00c8e22dbb9bbd4b0121463241fcbae992e31bdb1c4171114a2a8
|
| 3 |
+
size 214329627
|
audio/test-00003.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:11f66b4a467533066505d6d5fe4b48712effd638c1f2e78f30b503de72d29346
|
| 3 |
+
size 222528504
|
audio/test-00004.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e3fd46aebfd938a2431a59e2209cdbca1e05d8178d845f89c2023e20fe6da76
|
| 3 |
+
size 227270225
|
audio/test-00005.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b04efc3f846750b596faab943008a95f98202dff56608e5d217134c27d1350d4
|
| 3 |
+
size 225131607
|
audio/test-00006.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5a91d4e93caad6c3bfa09106ab6ca013d2f46475b88d7a842d22cf63fef477af
|
| 3 |
+
size 222435620
|
audio/test-00007.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:26d8d8156b956bb5da82a6597df201305c446582c6af143a61fbe678dfdb2871
|
| 3 |
+
size 223814415
|
audio/test-00008.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:767a5001fdd97f517898657fb6a9b11d4fa3bd2611b88f7e58b8c33a263f9b65
|
| 3 |
+
size 207566885
|
audio/test-00009.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6571efc0583376823533c01a743bde4b9c5d897391b120682670b5623f4aa555
|
| 3 |
+
size 198304443
|
audio/test-00010.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0fd4acaef9c8cfd10765c698883282cefc6cdc1a4ff764be2adc1519a83d5ad3
|
| 3 |
+
size 212547089
|
audio/test-00011.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:92f63102d444f921033865e4659295566fbf60ea888b595813f951e405629249
|
| 3 |
+
size 232135280
|
audio/test-00012.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:176d24b7882ffc7153aaca26e82e29f34009bfbb70fb6f44b70549e1b38d70dd
|
| 3 |
+
size 239392736
|
audio/test-00013.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b548934ccff9a333c24d9b0c21cfc3ea36ebbb94dbae5d85df1aed7435313afb
|
| 3 |
+
size 254305684
|
audio/test-00014.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:38da447af13f00305413e3a452da4f0459dde6eb933bf0795cd9dc1dd5feb6e9
|
| 3 |
+
size 255274708
|
audio/test-00015.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c3ee58c47ed4a2e16b3c26d25e1df297bad105b81c2b6c52d025a098de65d402
|
| 3 |
+
size 255090164
|
audio/test-00016.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c5af005869240a6b4a4033018d99f7825587f9b41168d8be9c95efe79fe5169c
|
| 3 |
+
size 255114216
|
audio/test-00017.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fbdcc16867b9f66f4adbbb576d602c6c88e6e20a2ab301849312da52fe71de08
|
| 3 |
+
size 229663852
|
audio/test-00018.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:55dbde644869c319080bc602c19c9c8e94642010065dfc89ca1cfb390eb98eab
|
| 3 |
+
size 232300697
|
audio/test-00019.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:122f1983dd2818732d30664b88d41029161c05e141a2ae11886b768d81ad2be0
|
| 3 |
+
size 157916239
|
audio/test-00020.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5ff1e1c41b4e95fa480c3591fff8fb668792bd766fd7975a28e813efc09bfcad
|
| 3 |
+
size 142570332
|
audio/test-00021.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:60addbf9f523824f36df2fa0147bedbd8f303f4cbd95a516ea0d57bf078a968c
|
| 3 |
+
size 141000255
|
audio/test-00022.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:52d2f7e54ebbae8b8c174917a7a8f212469e11991b74f82735ecd80198036270
|
| 3 |
+
size 196359460
|
audio/test-00023.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b1d043ac519ae6e60cca6ad7df91ebe13efc1b6862741617747acf773270e801
|
| 3 |
+
size 253394072
|
audio/test-00024.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:697caaf2d7abfa667a3d02a6a4149f0676e1202e374c52698edac6b49633128d
|
| 3 |
+
size 253371651
|
audio/test-00025.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fe57fb980fdc610fbd92ab43bf18e8d0ff418af143efb66c71a9c7537c8bb443
|
| 3 |
+
size 253190453
|
audio/test-00026.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:57c4e979a93e51dfe95e45f5b82e92f98ec98751bc14d8fd94475a159bd9e436
|
| 3 |
+
size 253089077
|
audio/test-00027.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:72dda13cbad4f9f75e4142991c7b5e7279bd2fd23b90bd44082f2ec4235f4dcc
|
| 3 |
+
size 253453654
|
audio/test-00028.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:588bba156d5d5e52362bebccb2b2782e995d276e3a0ebb8038060ae3285c1a20
|
| 3 |
+
size 253380871
|
audio/test-00029.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a17b73db9af8991dde9cdc8e0f54e9cc1e5b405297ff1e3bbfc77f30117e21d4
|
| 3 |
+
size 253043178
|
audio/test-00030.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3a39f685da32242a227a91d7bf08269066e4d474f4d038586aa162cec34ffdc0
|
| 3 |
+
size 253124918
|
audio/test-00031.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:13cad323510ef54cadd08fc89f4451bf7338b9d1042aa70312a0de44f4f89288
|
| 3 |
+
size 253355410
|
audio/test-00032.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d7bacae8780122ce817cef57202ca1583634f264003f8fb404cd19a65b6922a2
|
| 3 |
+
size 252929419
|
audio/test-00033.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0b9d10de5d1b205da93c4dc14717f2c2da7ffa8b7f1dc43df333e48c3b215ed1
|
| 3 |
+
size 253195848
|
audio/test-00034.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ff400c90c16a3949c6434d476d57ff5e19bd12e9f0549bccfa94ee195049e68f
|
| 3 |
+
size 252949819
|
audio/test-00035.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:36fb9eda30584438400374de62b624a18ab6f0e1b8b15032c1c4e88ecffd5a6b
|
| 3 |
+
size 252938386
|
audio/test-00036.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3f8e67d6db86144115eba60c158bd97d147f3200402ad9a90e207dceb26e4c96
|
| 3 |
+
size 252814908
|
audio/test-00037.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0a47cdd2221340580b30229c6efa5577620198e4e081d74a48233e6b6a2cb2b8
|
| 3 |
+
size 253470829
|
audio/test-00038.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eaee1b8bdd248d8197e0d641fa1056314a4f8400a8e74a013bb9bc500e25a910
|
| 3 |
+
size 253384053
|
audio/test-00039.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:02f595f609314428775339e320159f0dbfb847d5f1b68c377ad1b63080c076fa
|
| 3 |
+
size 253100212
|
audio/test-00040.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ca612366a862bddec117685d09f2ca0001974b6d7e2ee958890502a1841d9217
|
| 3 |
+
size 252740168
|
audio/test-00041.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9373af73e666eeb74e7fda199bfd00b291efbe931936d4ef4f66ffbd5e0febd0
|
| 3 |
+
size 253073923
|
audio/test-00042.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ddb9e6be8bd1c4cc6ffcb084a99ac2f29e01b8ea5698bbf1f951391b543112f8
|
| 3 |
+
size 252803864
|
audio/test-00043.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:87980669a0ca5ecf186676db351c21447f363d57e39d30a1f11851c3167bf99b
|
| 3 |
+
size 252584357
|