Dataset Viewer
Auto-converted to Parquet Duplicate
audio
audioduration (s)
15.5
81
dialogue_id
stringlengths
2
5
duration_seconds
float64
15.5
81
sample_rate
int64
24k
24k
audio_sha256
stringlengths
64
64
synthetic
bool
1 class
engine
stringclasses
1 value
role_0_voice
stringclasses
4 values
role_1_voice
stringclasses
4 values
14
44.8
24,000
a2779fbe6db2b012f2fa58fab887e6c293e269a0042e2d3a701b8ed40ace41b5
true
MoonCast
ears-p015
ears-p016
23
42.22
24,000
2351217c79be3897a4f2872bcdc1f9bee629c6e97879afd4870ab6383768b4c3
true
MoonCast
ears-p016
expresso-ex01
35
34.52
24,000
e044d2a1492615c2c17a5b3b8f07ec97cac7f84f99554bd58a1016477385e1d5
true
MoonCast
ears-p016
expresso-ex01
43
67.16
24,000
f782fa356c9783cc240ecef9e917b792b8341aec62c03086e205d01de643e8c6
true
MoonCast
expresso-ex01
ears-p015
71
30.78
24,000
95874ccd427be2bc58efac06628c0641ca2a2e916a38f1369cf0222a15b33962
true
MoonCast
expresso-ex01
ears-p015
84
49.34
24,000
8f18d7860f2c808b9584c0ad58ea2b12e2f85a79e2084259cf532c64fa7a1672
true
MoonCast
ears-p015
alba
90
60.78
24,000
79e7276b0965ecdc9a6054cbafc3b4164ceb1a7b98725b0e765ca5a3ac2e9920
true
MoonCast
ears-p015
alba
108
45.32
24,000
5ba366240d5b0397dc1c5e71670ed53afabe05b3579d5e26a81e4736b87f2c87
true
MoonCast
alba
ears-p016
114
54.64
24,000
7d1928bcdd41ef5cf1201302865ee345f76922fdef1ff8fce8ca4b04d534124e
true
MoonCast
expresso-ex01
ears-p015
118
35.7
24,000
886fc3edbdc91a4c6dba1c5d8b43ef0a1f9d869f5cf0a034d0c0b75ae55dfc2b
true
MoonCast
ears-p016
ears-p015
120
42.72
24,000
1fcda3cd55f296b9d7414ac6e4b726bf2da5a02b7235c1e4ed905819de3e709a
true
MoonCast
expresso-ex01
ears-p015
133
35.26
24,000
0451ce6f6be534785095dd4ab28ee396ad1359d9647978104ff5bdeb281d0550
true
MoonCast
alba
ears-p016
154
33.58
24,000
a4f48e4393bfd709ad540564706f290f2567d30867fcc6aff35842c38f82d336
true
MoonCast
alba
expresso-ex01
169
31.92
24,000
1abf63b13610706c7a1541b5418a2737051f7f97732d96829b672ac9975fa8fa
true
MoonCast
ears-p016
expresso-ex01
200
40.82
24,000
b02fcc4caff777ecd9a296759cc6bcf1716f5d029b0b26c0d67daa27ef1e79b5
true
MoonCast
ears-p016
ears-p015
201
40.6
24,000
0a48c3a0154e50a3f1c490b938b1c6bcb1ad29ae2613325b008eb746ccd4e1e2
true
MoonCast
expresso-ex01
ears-p016
212
48.42
24,000
1ab449f3c83b1d69abb13519be992a6fd3fcb9bb8cd29cb9eec819567d83a45b
true
MoonCast
ears-p015
expresso-ex01
228
42.66
24,000
3eec4c74b870b489e1f58152fd3a5fadc454ff6288f3377f6b60a211e42b23fa
true
MoonCast
expresso-ex01
ears-p015
231
65.44
24,000
12a324f70f7e67c5dc28a1dd921d367389c9a2cf50a40cdc003d756e88515058
true
MoonCast
ears-p015
alba
239
39.44
24,000
ecc2679bc8ff572dbc65f95a3a872c8a2e0ba1e0a3e57ab58879641136aaa491
true
MoonCast
ears-p016
ears-p015
247
37.86
24,000
b0020c5e0a55b9da40abd356b673e8474ada0ac1dac85597a7dcf1968631b26c
true
MoonCast
expresso-ex01
ears-p015
265
23.92
24,000
2ea3bda155db36a4eb1244bb17bb72bd24680f3e1ac4b9e6147793bc82c98b7d
true
MoonCast
alba
ears-p016
270
26.8
24,000
6f10e0e8ff24d94c6acce6656d563aeef2405988fcc23237de1fa8b77041ec05
true
MoonCast
expresso-ex01
ears-p016
273
53.88
24,000
4863a10a22b4e31b5284e9293efdab0d528b687d57d239abe3ee0ef7f17e97a6
true
MoonCast
ears-p015
expresso-ex01
285
41.3
24,000
3c1a7162443a523bcca1f651d857fe65ab7b86fa4be07ff283e166d22aba1fed
true
MoonCast
expresso-ex01
ears-p016
298
46.46
24,000
b789d27a307bdb021b1074918af03a05e851f9bc23f0f7b87f1cd283407f7ba1
true
MoonCast
ears-p015
expresso-ex01
304
35.6
24,000
ee7b5361cb9d4b7115b8ac3cca71e891209c4f008844ceabdfb4e6bbb14fabf8
true
MoonCast
ears-p016
expresso-ex01
342
47.92
24,000
b9bf2ffe1d4942d7eb82b09070dd2f3e1aa0696d2d228d3f4445707c0633957a
true
MoonCast
expresso-ex01
alba
348
24.82
24,000
2a94b77ffb4b8f52dc4ba95831efe431d13ab6a4e1d72b6be46dac345f847abc
true
MoonCast
alba
expresso-ex01
356
19.6
24,000
a4ddd57e8335bb409a1bfc3953493e5570c76c486ca349be7d520aa63dac96db
true
MoonCast
alba
expresso-ex01
365
38.86
24,000
9a19d1dcccf8573cd3be2209028bce84f45d780b8b2bc7bc0b5152f35fc5dd4f
true
MoonCast
ears-p015
alba
373
46.58
24,000
92fd64954c47fe42ad4f2cf34c0651387943ca383f523d694e55f9368f454875
true
MoonCast
ears-p016
expresso-ex01
374
45.88
24,000
fbba10c33287bd66c940661e52b0e3b66f9924c0fb5444ed00afa6666af1853e
true
MoonCast
alba
ears-p016
390
48.26
24,000
7d31806725d4d08997a49e2edc3b967484eacd757bd396d2288227292d92a5e8
true
MoonCast
expresso-ex01
ears-p015
396
40.86
24,000
3f76791dea1fd54f0f4c86353b467bde9f0dd60584f3333f05026ad596ae11b3
true
MoonCast
expresso-ex01
ears-p015
414
58.76
24,000
d187c43dcdacc6229e82b186a8943f91cab10cb52a573f06aee8a630ca8e4dd3
true
MoonCast
alba
expresso-ex01
420
28.3
24,000
b32044fa6a6aea7754c6fe57b370f8342d967b99b03694278d46b935d083a76e
true
MoonCast
ears-p016
expresso-ex01
438
31.88
24,000
76767003fd1df244de0348d5f01f3ce6abed378d8efc0ae453e9da635140b372
true
MoonCast
ears-p016
alba
439
58.32
24,000
4419460535f3547411e8b57d6d6a862186ae0f480f8db7ba894ece8817e90b49
true
MoonCast
ears-p015
ears-p016
440
38.6
24,000
d9300d22610bc00a2d7f1501c0a25fe89bbc7bf99fbdd445ccac8dc746af3a97
true
MoonCast
ears-p015
expresso-ex01
466
39.46
24,000
aeaadd97f1354caf213cf6ff28fa839923f80104e6249073f0ae77cd52c8d192
true
MoonCast
expresso-ex01
ears-p016
479
37.82
24,000
6ff3a44fddaa01bb73f4528f0872c44d824c94b291d13d4b1e51da81cf502c60
true
MoonCast
ears-p016
alba
484
42.2
24,000
e13916dc8af2f8be4524225e97399d67a4213f320b5422885e1911088991172d
true
MoonCast
ears-p016
ears-p015
503
24.58
24,000
7dd626ec56fc0027e273aa4d26ed36182f3528402c79692e906360964802d763
true
MoonCast
alba
ears-p016
527
38
24,000
3f9eb331d1cbda1c8bb571b973c1822344b7d4fcf422e7bda9c14d856bf37226
true
MoonCast
ears-p015
ears-p016
539
25.44
24,000
76fb3ad05f0ffa1b93b8f78b421a0828f908b179374b3ea0ee9ec4bcc4e56c46
true
MoonCast
expresso-ex01
alba
558
30.88
24,000
64b8855bd54d78ab7775a07505a2d2d5272c6e7df4b0defaf0dff9c294b34413
true
MoonCast
ears-p016
alba
581
44.56
24,000
dc274101aff5758fdc94bff0d020c85c9f8fcddcdef021441c3e9b2c951ae8a6
true
MoonCast
ears-p015
expresso-ex01
597
57.12
24,000
872208f7453419d8e3396b7a8642a7b2a55261293cace6cbcb53924f4335d178
true
MoonCast
ears-p015
alba
600
33.68
24,000
8742b964c0bc6f19ff8695044e2d4e97b400a0170d2f36747a32cdfeaf3b3073
true
MoonCast
ears-p016
ears-p015
608
33.06
24,000
6eea053bde8864a688174d286a7d6b51bf0666d5c0522ab192ade40cca588dd6
true
MoonCast
alba
expresso-ex01
630
37.82
24,000
b4d160e67be77040c5676a196bf5b91e9e27f514544e7aaf4de34817153869db
true
MoonCast
ears-p015
expresso-ex01
638
39.62
24,000
20c4b83ef62e3d4c26fef219b44db7fc35beedf9d61fbe52b6d7105a8dcb35b6
true
MoonCast
ears-p015
ears-p016
639
32.84
24,000
7b26772294b95e1c1e5e5bea4a999ca98dfcef377740f0bae4c36852ff636367
true
MoonCast
expresso-ex01
ears-p015
649
36.02
24,000
6fa5650e53dd0f5ded587272436bc8b22913a1baf3d20ac56c312ac278f5c53b
true
MoonCast
ears-p016
expresso-ex01
671
26.76
24,000
290fae38973d7353b7c25965b6de7b811160b7cbf760699ee42e0580a0a15614
true
MoonCast
ears-p016
alba
695
28.42
24,000
5cef78680188dc04676f66e2c0bffe20d5cad637102be8fe02fb7112461b3171
true
MoonCast
ears-p016
expresso-ex01
706
32.08
24,000
7bd6ea9cf0cdaebcc689947e235b7a19c9c9de03041d5e91f20092f3e871c84e
true
MoonCast
alba
ears-p015
719
44.2
24,000
eacc17701728c459bb2b78c1337af620b25c22fe0be9601c67b82653cc8190de
true
MoonCast
ears-p016
alba
722
34.54
24,000
45af26aa8f0df7341f9b66b215f7088b898c2a2ebe68031f7e6a991e67708a7b
true
MoonCast
ears-p016
alba
773
30.76
24,000
331f23648eb02951ab7a60855ee90d3cc94ccbb32ac0a5121088c1cea3402823
true
MoonCast
ears-p016
ears-p015
787
35.44
24,000
0965f53636e78309cb2ba4d3f9e77d2c37eb20410a396fc4cf5ce344d7dca493
true
MoonCast
expresso-ex01
ears-p015
802
46.18
24,000
60a8fdc644bb69ef94d31ddd2dbb552a06954ac2b81a8837ab1db68f4ca43749
true
MoonCast
alba
ears-p016
821
36.18
24,000
2ec5f756970531e9d7e1339e5bdf7b4141d1aafd8bd8d716f62c8737ee0fe05f
true
MoonCast
ears-p016
ears-p015
823
30.06
24,000
a354ac3c8175f54488a34733ae74338a86840a6de7d0c3d9067990c455f0ebae
true
MoonCast
ears-p016
ears-p015
909
36.7
24,000
3a36c16e64a7e72913fca2a0dbd4630c3bd692c876394ccb5d0dd9f8bf284050
true
MoonCast
ears-p015
ears-p016
923
38.94
24,000
0822264c079943fa3e0da845f0887c28817aa3ea54a26f0627bb5fe6cdbd50b2
true
MoonCast
ears-p015
ears-p016
931
33.58
24,000
3e538c17f0058d21ded76a1e4aecd6748aec8a2c894e389cd3b82ba79377e8ec
true
MoonCast
ears-p016
expresso-ex01
938
37.66
24,000
eeee7a46dcecca91b73af87d70b086558d0e84871531b9fcbf37147a2895f7e2
true
MoonCast
ears-p016
alba
958
52.92
24,000
c723eb6e83be94d4dd28f68f99462286c78e54eb3a2bbdf446fdce8f719f4da3
true
MoonCast
expresso-ex01
ears-p015
970
63.78
24,000
35e32ad4e22567e0dba0a779d224dd8b3e6b57a9dcbef63662d0a14ef7fa40f7
true
MoonCast
ears-p016
ears-p015
975
38.72
24,000
ea40677aea1cfc9828b28e4e78ef0c51049a2dc952951545f9db2a8192356321
true
MoonCast
expresso-ex01
ears-p015
976
34.14
24,000
a1e819890ba37b1714246615b867c935f0db32e2fbc7642895a1f47f30627e3f
true
MoonCast
expresso-ex01
ears-p015
997
29.64
24,000
080259f6d0a01910a342467671fdef6d440686e404a288d333ed412c64d61227
true
MoonCast
ears-p016
expresso-ex01
1027
29.02
24,000
cafb415f1edef42b577bdeae2c4aef3f15ceb3bcb5822319ac9ca2392a1b9fb9
true
MoonCast
expresso-ex01
alba
1041
64.26
24,000
e0449ad5be8babbd659783aa2bbcbb8e94f27b65270ab27dc7fe7666ec1ab2b3
true
MoonCast
expresso-ex01
ears-p015
1044
50.16
24,000
3398cdcffa147e010b939a14bdebd26050842d23f6a98e664c8827821f0a6057
true
MoonCast
expresso-ex01
ears-p015
1047
26.8
24,000
f56264ca5df01ba529c68eeb2265c8ae09a6c9153d1395ccc7fdb06766f54a67
true
MoonCast
ears-p016
alba
1069
24.86
24,000
02f0ce8277e44b3c1edbf93f6598d7497ad6b644cf38de56620b160c7119fb5c
true
MoonCast
expresso-ex01
ears-p016
1107
39.68
24,000
bf629ec9c0fb2447d246d43ff032aa24675ab4023b5aa8d4a2c61c0b6fee1145
true
MoonCast
expresso-ex01
ears-p016
1131
26.5
24,000
ac09d33cb6518889e9355f3850cf33a2b4ca28121ff178d56a448278a661a4fd
true
MoonCast
ears-p015
alba
1135
32.22
24,000
354c783a7167ff9d0b9b437a1b89fc48c049fc095bc88a66a129e19565ca7b28
true
MoonCast
expresso-ex01
alba
1143
36.42
24,000
dc0659cbacf0faa7244a55f7b550fd46d2339112cc8b387e62eff445860734c0
true
MoonCast
expresso-ex01
ears-p015
1144
45.44
24,000
ef2c363b14dd16b822c5ac5033ff763fa71c7b0c168dda6ac3b7212c64d8bb0d
true
MoonCast
ears-p015
alba
1161
48.26
24,000
1e71db12e473c2cec0501cd15bd7b95a48a208a3e5df7dd5bd40ae28c5518a28
true
MoonCast
ears-p015
expresso-ex01
1170
47
24,000
eb1662ef148c7cdea8be3e1cc2c85f53d090586971ed675ee740597123fc4c9a
true
MoonCast
ears-p016
alba
1174
29.38
24,000
51eddecd69f38a08d3af59404637293a1f87dc8baeba4670130783066d093a37
true
MoonCast
expresso-ex01
alba
1177
35.98
24,000
cab69bde9084e4d8b96c552feba0d2201ebe7c07ec21fd1eecffd9ce187d9e23
true
MoonCast
ears-p016
expresso-ex01
1180
36.84
24,000
abcf750a40b700c6e08bf3b6c444398b79359724e9bac15dc8b0edc42ef5cf92
true
MoonCast
ears-p015
ears-p016
1181
35.1
24,000
9ed1d032bd08257ea99c0689ee842071293d051e03fa8e15554f2614be1dab9d
true
MoonCast
alba
ears-p016
1198
50.12
24,000
46e1f50537a272738e15e6f54207ca20140528707cf4d592918dccb79b6b7075
true
MoonCast
ears-p015
expresso-ex01
1204
31.88
24,000
ceab12f453c1feea6e5bd029d4e9df3f5bd39cdf739803342898054bc68d8331
true
MoonCast
ears-p016
ears-p015
1205
41.4
24,000
5c9f2ccdb621cb9702caded180733c783e1eb2e57a3a5a0fdcb63cf8f7a9bad0
true
MoonCast
ears-p016
ears-p015
1213
31.5
24,000
2f30242cd5c547b8398515469dc676ec7b360f691c26a829bce23e39785e3c7f
true
MoonCast
ears-p015
alba
1260
40.06
24,000
68316aec8c7bab6bc18f1eb7beb208c3309dc88de50c35dba47cbaa569199cb1
true
MoonCast
ears-p015
ears-p016
1279
31.8
24,000
9462cf32149e183152f555094239a021528814533cfecb9d259bedf2aff15eef
true
MoonCast
ears-p016
expresso-ex01
1281
36.74
24,000
25b59a17aad34c05b207784b9ef2a0358a6c0575b45251b0c8f5358618cbf76e
true
MoonCast
ears-p015
ears-p016
1291
60.04
24,000
cbcb0a16ef4ab7fa9e929d9494d998d503e5a49aff075ddf98e549e6d11037b1
true
MoonCast
alba
ears-p016
1300
46.8
24,000
1a6260a10ab28c826f62d4d82eefec72e9f74c3003a7d6cd743ddce32132fe67
true
MoonCast
expresso-ex01
alba
1312
33.98
24,000
06836eaec97349653cb72a3c516ae8aef5fadcc50ace0181b313105b25c5061f
true
MoonCast
ears-p015
alba
End of preview. Expand in Data Studio

MAD2

MAD2 contains 1,000 synthetic English conversations, their original written dialogues, ASR transcripts of the replacement recordings, and current sentence check-worthiness labels. Audio was generated with MoonCast using alba, ears-p015, ears-p016, and expresso-ex01 reference voices.

The package contains 11.04 hours of mono 24 kHz PCM16 audio and 8,192 labeled sentences: 1,230 check-worthy and 6,962 negative. Each dialogue has two distinct voices; each reference voice occurs in 500 dialogues.

This is the compact local release package, not a record of public-release approval. Sentence forced alignment is not included: sentence start/end times remain null. The dataset release version remains to be finalized. The paper citation is provided below.

Audio Release and Re-Synthesis

The paper's experiments and WER evaluation use the original recordings generated with MoonCast's two default demonstration voices. MoonCast's speech-generation notice prohibits redistribution of its bundled demonstration audio prompts and audio generated from those prompts. We therefore regenerated the recordings for public release using the four alternative reference voices listed above. The original default-voice recordings are not distributed in this package.

The released audio is a replacement corpus: the paper's reported model performance and WER do not evaluate these replacement recordings.

Researchers are welcome to use the scripts in dialogues/original.jsonl to generate their own audio with other TTS systems and voices for noncommercial research, subject to the MAD2 license, applicable third-party terms, and the permissions required by their chosen TTS models and voices. Please identify the synthesis system and voices used and distinguish any new recordings from the supplied MAD2 audio. New recordings require their own ASR transcripts and timing alignments; timestamps from the supplied audio should not be reused.

Files

MAD2_release/
  README.md
  LICENSE
  THIRD_PARTY_NOTICES.md
  VOICE_ATTRIBUTION.md
  licenses/
  dialogues/
    original.jsonl
  audio/
    <dialogue_id>.wav
    metadata.jsonl
  asr/
    transcripts.jsonl
  annotations/
    labels.jsonl

The four JSONL files each contain 1,000 rows, joined by the string dialogue_id. There is one WAV per dialogue. JSONL is UTF-8 with one JSON object per line. No train/dev/test partition is prescribed; corpus names the complete collection, not a training split. Researchers must specify their own evaluation partition and keep all sentences from a dialogue together when splitting.

Original Dialogues

dialogues/original.jsonl stores the written dialogue text once, preserving the sentence strings, speaker roles and turn boundaries from the hash-verified server scripts. It does not include the old script labels. Fields:

  • dialogue_id, source_dataset, source_claim_id: corpus and LIAR source identity.
  • scenario_type, spread_style: corrected server categories.
  • source_script_sha256: hash of the original source JSON, not this packaged row.
  • turns: original_turn_idx, role, and sentences. Each sentence has sentence_id, original_sent_idx, and text.
  • audio_script: compact text edits and turn spans that map the original dialogue to the exact text used to synthesize the released recording, without storing another full dialogue copy.

Indices are zero-based. A sentence ID is <dialogue_id>:<original_turn_idx>:<original_sent_idx>. Roles 0 and 1 refer to the voice assignments in audio metadata, not automatically inferred diarizer IDs.

Audio-Script Differences

Nine dialogues have text edits: seven added acknowledgments and two apostrophe corrections. Twelve inputs merge adjacent same-role turns. The original text and current sentence IDs are preserved; generated and original turn indices differ.

To interpret audio_script, first join the original sentence strings in order with one ASCII space. text_edits contains ordered, nonoverlapping char_start, char_end, replacement patches against that string. Apply them from last to first. The resulting string is the exact joined synthesis text. Then each turn_spans entry selects a generated turn using generated_turn_idx, role, char_start, and char_end in the resulting string.

All character spans are half-open Python Unicode-character offsets, not UTF-8 byte offsets. These are text mappings, not audio timestamps.

Audio

audio/metadata.jsonl contains file_name, dialogue_id, duration_seconds, sample_rate, audio_sha256, synthetic, engine, role_0_voice, and role_1_voice. file_name is relative to the audio directory. No full script, ASR transcript, or sentence-label table is duplicated in this metadata.

All 1,000 WAVs are the unchanged approved replacement takes. 979 passed automatic QA; the remaining 21 were accepted by the dataset owner, including seven exact takes reviewed by listening and 14 accepted as minor differences. This acceptance does not mean every ASR word or diarization boundary is correct.

ASR

asr/transcripts.jsonl contains dialogue_id, the recognized transcript, and words. Word entries preserve word, start, end, and probability from the new-audio faster-whisper large-v3 reports. Times are seconds relative to the WAV. The transcript and word sequence are two views of ASR output, not corrected script text. Word times/confidences are automatic, not human gold annotations; words are not assigned speaker identities in this table.

Labels

annotations/labels.jsonl stores one row per dialogue with:

  • source_claim_veracity: inherited binary LIAR seed label (true or false), not a truth judgment about every utterance.
  • source_claim_original_label: original LIAR true, false, or pants-fire. false and pants-fire map to binary false; true remains true.
  • annotation_version: server_active_aligned_20260913, identifying the current server label layer. The name does not assert alignment to the new recordings.
  • sentences: sentence_id, check_worthy, generated_text_spans, audio_start, audio_end, and alignment_status.
  • audio_only_turns: the seven synthesis-only acknowledgments, identified by generated_turn_idx, with check_worthy: null and annotation_status: synthesis_only_addition_unlabeled.

check_worthy is a selection label, not factual correctness. The current 1,230 positive labels are retained, not the old 3,368 positives. Current does not imply human-verified. Audio-only acknowledgments are not counted among the 8,192 labeled sentences and must not silently be scored as negative.

generated_text_spans selects text within the reconstructed generated turn: generated_turn_idx, char_start, and char_end. Matching ignores case and punctuation, enforces the speaker role, and can omit boundary punctuation. It is not forced alignment. audio_start and audio_end are null, with status pending_new_audio_sentence_alignment; no old-audio sentence times are reused.

Loading

The default AudioFolder view loads audio and minimal metadata only. Load the other tables separately and join on dialogue_id; sentence text and labels join on sentence_id.

from datasets import load_dataset

audio = load_dataset("./MAD2_release", split="corpus")
dialogues = load_dataset("json", data_files={
    "corpus": "./MAD2_release/dialogues/original.jsonl"
}, split="corpus")
asr = load_dataset("json", data_files={
    "corpus": "./MAD2_release/asr/transcripts.jsonl"
}, split="corpus")
labels = load_dataset("json", data_files={
    "corpus": "./MAD2_release/annotations/labels.jsonl"
}, split="corpus")

No custom loader or experimental code is required in the repository. Local validation was performed with datasets 3.6.0; decoding dependencies depend on the installed library version.

Scope and Limitations

MAD2 supports research on conversational check-worthiness and misinformation. These are synthetic dialogues, not authentic conversations or evidence that a reference performer made the statements. Content can be false, political, misleading or offensive. Inherited seed labels and automatic annotations are not independent fact-checking evidence. The four-voice synthetic setting and source bias limit generalization. Earlier results on the original audio do not measure this replacement corpus.

Experimental splits, old labels, context summaries, diagnostic QA/diarization reports, generation logs, validation scripts, and release-preparation records are kept locally, outside this dataset package. Original restricted MoonCast audio, reference WAVs, LIAR TSVs, credentials, model weights and runtime code are not distributed here.

License and Attribution

MAD2-controlled contributions use the MAD2 Noncommercial Research License 1.0, permitting the specified noncommercial research uses only within the licensor's controlled rights. Upstream rights remain separate. Alba's source recording is CC BY 4.0; EARS and Expresso references are CC BY-NC 4.0; LIAR retains its source-copyright and research-use notice. Neither the custom license nor this repackaging certifies public cloned-voice scope, source/service compliance, or authority over any institutional/collaborator rights. Future license revisions do not retroactively replace validly granted permissions.

See source and method notices, voice attribution, and the unchanged texts in licenses/. LIAR: William Yang Wang (2017), "Liar, Liar Pants on Fire": A New Benchmark Dataset for Fake News Detection. Synthesis: MoonCast.

Contact/licensor for controlled or authorized rights: Chaewan Chun, czc5884@psu.edu.

Citation

@inproceedings{chun-etal-2026-context-aware,
  author    = {Chun, Chaewan and Zhang, Delvin Ce and Lee, Dongwon},
  title     = {Context-Aware Multimodal Claim Verification in Spoken Dialogues},
  booktitle = {The 2nd Speech and Audio Language Models Workshop
               ({SALMA}), {EMNLP}},
  year      = {2026}
}
Downloads last month
72