2025:Audio Beat Tracking Results - Revision history

Junyan: /* Submissions */

2025-09-19T14:20:06Z

‎Submissions

2025-09-17T06:49:21Z

‎Submissions

2025-09-16T09:33:01Z

2025-09-16T04:29:21Z

2025-09-14T05:09:08Z

‎SMC

2025-09-14T03:30:17Z

2025-09-13T06:04:44Z

‎Test Sets

2025-09-13T05:57:03Z

‎SMC

2025-09-13T05:56:27Z

‎Submissions

2025-09-13T05:56:00Z

‎GTZAN

@@ Line 21: / Line 21: @@
 | KG-ApolloBeats
 | The 2025 KG Music Beats Tracking System
-| TBA
+| [http://futuremirex.com/portal/wp-content/uploads/2025/audio-beat-tracking/KG-ApolloBeats.pdf PDF]
 | DingKun Xiao, Haijun Cai, Chuanyi Chen
 |-
 | KG-ApolloBeats 2
 | The 2025 KG Music Beats Tracking System
-| TBA
+| [http://futuremirex.com/portal/wp-content/uploads/2025/audio-beat-tracking/KG-ApolloBeats.pdf PDF]
 | DingKun Xiao, Haijun Cai, Chuanyi Chen
 |-
 | BeatU
 | Beat-U: Multi-Task Music Understanding with Hierarchical Timescales
-| TBA
+| [http://futuremirex.com/portal/wp-content/uploads/2025/audio-beat-tracking/BeatU.pdf PDF]
 | Jingwei Zhao, Rei Nishiyama, Kouhei Sumi, Takuya Fujishima, Akira Maezawa
 |}

@@ Line 32: / Line 32: @@
 | Beat-U: Multi-Task Music Understanding with Hierarchical Timescales
 | TBA
-| YAMAHA Corporation
+| Jingwei Zhao, Rei Nishiyama, Kouhei Sumi, Takuya Fujishima, Akira Maezawa
 |}

@@ Line 1: / Line 1: @@
 == Submissions ==
 {| class="wikitable"
@@ Line 26: / Line 24: @@
 | DingKun Xiao, Haijun Cai, Chuanyi Chen
 |-
-| KG2
+| KG-ApolloBeats 2
-| Same as KG-ApolloBeats, but not trained on SMC or GTZAN
+| The 2025 KG Music Beats Tracking System
 | TBA
 | DingKun Xiao, Haijun Cai, Chuanyi Chen
@@ Line 77: / Line 75: @@
 | 92.77
 |- style="vertical-align:bottom;"
-| style="text-align:left;" | KG2
+| style="text-align:left;" | KG-ApolloBeats 2
 | 88.21
 | 76.45
@@ Line 147: / Line 145: @@
 | 76.22
 |- style="vertical-align:bottom;"
-| style="text-align:left;" | KG2
+| style="text-align:left;" | KG-ApolloBeats 2
 | 58.55
 | 45.08
@@ Line 213: / Line 211: @@
 | 91.79
 |- style="vertical-align:bottom;"
-| style="text-align:left;" | KG2
+| style="text-align:left;" | KG-ApolloBeats 2
 | 92.98
 | 82.75
@@ Line 278: / Line 276: @@
 | 93.48
 |- style="vertical-align:bottom;"
-| style="text-align:left;" | KG2
+| style="text-align:left;" | KG-ApolloBeats 2
 | 95.87
 | 87.43

@@ Line 212: / Line 212: @@
 | 82.26
 | 91.79
 |- style="vertical-align:bottom;"
 | style="text-align:left;" | Baseline: BeatThis
@@ Line 267: / Line 277: @@
 | 84.72
 | 93.48
 |- style="vertical-align:bottom;"
 | style="text-align:left;" | Baseline: BeatThis

@@ Line 157: / Line 157: @@
 | 61.45
 |- style="vertical-align:bottom;"
-| style="text-align:left;" | Baseline: BeatThis
+| style="text-align:left;" | Baseline: BeatThis*
 | 71.81
 | 55.64

@@ Line 23: / Line 23: @@
 | KG-ApolloBeats
 | The 2025 KG Music Beats Tracking System
 | TBA
 | DingKun Xiao, Haijun Cai, Chuanyi Chen
@@ Line 38: / Line 43: @@
 * '''Yamaha_JPOP''': A private dataset annotated by Yamaha Corporation. The dataset contains 250 JPOP songs.
 * '''Yamaha_Balanced''': A private dataset annotated by Yamaha Corporation. The dataset contains 239 songs. While it is still biased towards JPOP songs, the dataset covers a wider range of genres: J.Pop (10.37%), Rock (10.37%), J.Enka (10.37%), J.Kayoukyoku (10.37%), Soundtrack (10.37%), Western Pop (10.37%), Children's Song (10.37%), R&B (6.22%), Hiphop (4.56%), Jazz (2.49%), Dance (2.49%), World (2.07%), Techno (1.24%), Easy listening (1.24%), J.Minyou (1.24%), Others (5.81%).
 == GTZAN ==
@@ Line 72: / Line 76: @@
 | 86.86
 | 92.77
 |- style="vertical-align:bottom;"
 | style="text-align:left;" | Baseline: BeatThis
@@ Line 93: / Line 107: @@
 | 86.69
 |}
 Entries with [*] are trained on this dataset.
@@ Line 112: / Line 127: @@
 ! style="vertical-align:bottom;" | AMLt
 |- style="vertical-align:bottom;"
-| style="text-align:left;" | BeatU*
+| style="text-align:left;" | BeatU
 | 53.14
 | 40.67
@@ Line 132: / Line 147: @@
 | 76.22
 |- style="vertical-align:bottom;"
-| style="text-align:left;" | Baseline: BeatThis*
+| style="text-align:left;" | KG2
 | 71.81
 | 55.64
@@ Line 152: / Line 177: @@
 | 29.48
 |}
 Entries with [*] are trained on this dataset.

@@ Line 36: / Line 36: @@
 * '''GTZAN''': 999 songs from the GTZAN dataset (starting from next year, training on GTZAN will be disallowed)
 * '''SMC''': 217 songs from the SMC collection
-* '''Yamaha_JPOP''': A private dataset annotated by Yamaha Corporation. The dataset contains 200 JPOP songs.
+* '''Yamaha_JPOP''': A private dataset annotated by Yamaha Corporation. The dataset contains 250 JPOP songs.
-* '''Yamaha_Balanced''': A private dataset annotated by Yamaha Corporation. The dataset contains 241 songs. While it is still biased towards JPOP songs, the dataset covers a wider range of genres: J.Pop (10.37%), Rock (10.37%), J.Enka (10.37%), J.Kayoukyoku (10.37%), Soundtrack (10.37%), Western Pop (10.37%), Children's Song (10.37%), R&B (6.22%), Hiphop (4.56%), Jazz (2.49%), Dance (2.49%), World (2.07%), Techno (1.24%), Easy listening (1.24%), J.Minyou (1.24%), Others (5.81%).
+* '''Yamaha_Balanced''': A private dataset annotated by Yamaha Corporation. The dataset contains 239 songs. While it is still biased towards JPOP songs, the dataset covers a wider range of genres: J.Pop (10.37%), Rock (10.37%), J.Enka (10.37%), J.Kayoukyoku (10.37%), Soundtrack (10.37%), Western Pop (10.37%), Children's Song (10.37%), R&B (6.22%), Hiphop (4.56%), Jazz (2.49%), Dance (2.49%), World (2.07%), Techno (1.24%), Easy listening (1.24%), J.Minyou (1.24%), Others (5.81%).
 == GTZAN ==