diff --git a/.gitattributes b/.gitattributes
index 1d5e70fe6001f78c49063f34431e25c10bbf5816..10fa385ae216b79522d202fc15fc4a6c7d32fb4f 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -134,3 +134,103 @@ r1-a/dataset/gsm8k_with_audio/test/374.wav filter=lfs diff=lfs merge=lfs -text
 r1-a/dataset/gsm8k_with_audio/test/373.wav filter=lfs diff=lfs merge=lfs -text
 r1-a/dataset/gsm8k_with_audio/test/539.wav filter=lfs diff=lfs merge=lfs -text
 r1-a/dataset/gsm8k_with_audio/test/363.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/462.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/366.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/378.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/444.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/313.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/368.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/3.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/387.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/583.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/307.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/38.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/311.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/541.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/37.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/399.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/453.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/449.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/370.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/520.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/306.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/310.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/54.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/361.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/540.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/34.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/613.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/451.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/441.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/543.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/551.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/610.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/382.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/291.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/456.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/297.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/506.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/548.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/545.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/623.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/357.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/445.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/544.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/619.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/542.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/546.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/618.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/386.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/398.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/616.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/608.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/557.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/607.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/609.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/602.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/614.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/62.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/549.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/308.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/606.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/379.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/552.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1113.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1107.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1039.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1104.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/617.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1041.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1034.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/611.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/367.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/455.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/547.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1099.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/396.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1103.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1105.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/61.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/442.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/620.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1117.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1100.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1102.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/53.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/105.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1045.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/26.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1056.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/111.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1037.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1031.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1061.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1059.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1046.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1058.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1048.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1112.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1121.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/390.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1051.wav filter=lfs diff=lfs merge=lfs -text
+r1-a/dataset/gsm8k_with_audio/test/1062.wav filter=lfs diff=lfs merge=lfs -text
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1031.wav b/r1-a/dataset/gsm8k_with_audio/test/1031.wav
new file mode 100644
index 0000000000000000000000000000000000000000..87958b48fc3bd0fb0e1b1d232eff5504b022a536
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1031.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:503efc0f11c3b8d6ac99a39d5434b678972acabb162cedea8e6d7d106c69efe3
+size 1843280
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1034.wav b/r1-a/dataset/gsm8k_with_audio/test/1034.wav
new file mode 100644
index 0000000000000000000000000000000000000000..db2c3861c20737af0ae244ec49c4f1d069261a5b
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1034.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73e1b3b6fa0d6b2c9f9b3f3b8d0d809900c1f7dbf76ba4e4d52d2877965edab6
+size 1048400
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1037.wav b/r1-a/dataset/gsm8k_with_audio/test/1037.wav
new file mode 100644
index 0000000000000000000000000000000000000000..c91e8670b4ea3428d609efb106148a4bba7e453f
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1037.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:743c451e8b1d8a79f79e132e4964dc6901d0520f78091d5e65f0e23c1763d582
+size 1440080
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1039.wav b/r1-a/dataset/gsm8k_with_audio/test/1039.wav
new file mode 100644
index 0000000000000000000000000000000000000000..bb29cf368328d7ddde0355927896702a57cdc820
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1039.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c64cf10408f7b6f70b31d50dd14c7ab222ccc6dd4c22e776f919b8419d802587
+size 1228880
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1041.wav b/r1-a/dataset/gsm8k_with_audio/test/1041.wav
new file mode 100644
index 0000000000000000000000000000000000000000..90e659315e84d5d1818ed9a13a0008e3959636f8
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1041.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:762c08b1dbe2edc78185781ed5f3a8430c46b70d385c966a2be288963c3e8fc1
+size 1367120
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1045.wav b/r1-a/dataset/gsm8k_with_audio/test/1045.wav
new file mode 100644
index 0000000000000000000000000000000000000000..d4c7b45992690ae9e60d0f00e5f7173c7f675246
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1045.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38c706de7361555d27ae5b94835c80eadd7d7bab017b5987f3e80cdbcd151b78
+size 1140560
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1046.wav b/r1-a/dataset/gsm8k_with_audio/test/1046.wav
new file mode 100644
index 0000000000000000000000000000000000000000..f7c25a4047582146ae06652198db814159201b5a
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1046.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b091b433815f65ccb268006a8e56683de096781f4f7eb4400643ca9851672bf1
+size 929360
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1048.wav b/r1-a/dataset/gsm8k_with_audio/test/1048.wav
new file mode 100644
index 0000000000000000000000000000000000000000..b553422a979e191dfe4487681bf9abb02d8fefb8
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1048.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d1b6d360a384474e5dae0205a58b0c0c0c9d948bb0834409c01e0e51a76c5da
+size 1440080
diff --git a/r1-a/dataset/gsm8k_with_audio/test/105.wav b/r1-a/dataset/gsm8k_with_audio/test/105.wav
new file mode 100644
index 0000000000000000000000000000000000000000..886fb3d9e12aa60dcb309b7c3bbe6bdd748948cd
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/105.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e09218bb0a5a495b49adca3dcc2f68e3b99f7a0fcf3b0dc92d74235ecb7e42b8
+size 794960
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1051.wav b/r1-a/dataset/gsm8k_with_audio/test/1051.wav
new file mode 100644
index 0000000000000000000000000000000000000000..04eca95b1114a7a78c8975c9dc34d2cce7570e4c
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1051.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4673ecdd60ce168a2d6aececbaabaf42d76ddd32885da6edefec892456fa6e1b
+size 1040720
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1056.wav b/r1-a/dataset/gsm8k_with_audio/test/1056.wav
new file mode 100644
index 0000000000000000000000000000000000000000..d87e648c95bbce214a23c8704d2d24f5160f106f
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1056.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f88fc5e8abd9b516df54c3a12360a2e1e2596ed0ecb6ebfa342b16c27c23aa6
+size 933200
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1058.wav b/r1-a/dataset/gsm8k_with_audio/test/1058.wav
new file mode 100644
index 0000000000000000000000000000000000000000..7e76e9dc75ad1ce0504008c0e2549a5cd610b2d3
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1058.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5c408fb43cbd39eb0d0eb22c6490fccefb8621cc6dbcec76f8d6256ddf22592
+size 1209680
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1059.wav b/r1-a/dataset/gsm8k_with_audio/test/1059.wav
new file mode 100644
index 0000000000000000000000000000000000000000..aabff0376bee273fae74ccfd2848e934757e3484
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1059.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d30cab84993cc4b613b4d5b79ccf2809ff0bcfb941bff7e31dc2f0ba6545da0
+size 894800
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1061.wav b/r1-a/dataset/gsm8k_with_audio/test/1061.wav
new file mode 100644
index 0000000000000000000000000000000000000000..eba10a174737e59ffa1ab7757f7571dedf45ef8c
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1061.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb8d11533fff59b54b6dde0e304951be1646c77e10f1533c2137a948af5c65ea
+size 768080
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1062.wav b/r1-a/dataset/gsm8k_with_audio/test/1062.wav
new file mode 100644
index 0000000000000000000000000000000000000000..a8a7af3364a7e337e2f41b8612c451f7797f86ba
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1062.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ba120eb589bedc7ae21acb39b6b5b70c01040019226e8b43a97709969cf5ec8
+size 1824080
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1099.wav b/r1-a/dataset/gsm8k_with_audio/test/1099.wav
new file mode 100644
index 0000000000000000000000000000000000000000..2c4ca1f6a356d9957692ada696243e9a292fdc8e
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1099.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00ed191f00535323ff5c024190c0d92a044620e0c93044b45b135d7c3ce8f5f3
+size 1751120
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1100.wav b/r1-a/dataset/gsm8k_with_audio/test/1100.wav
new file mode 100644
index 0000000000000000000000000000000000000000..8d2f9d42074606abb760941a2cc9b99df056f391
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1100.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd31790651932b35266220c8d133d00abdddb07211716fbf0d7c26d9fe7f509c
+size 994640
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1102.wav b/r1-a/dataset/gsm8k_with_audio/test/1102.wav
new file mode 100644
index 0000000000000000000000000000000000000000..0d8693c2a1e5e1b7527554107df3092db67d089d
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1102.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e71a178dacd4509a0c9b0b20f4c9df53da20e88764fc0bc66b63c9f3ec99c0b
+size 1804880
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1103.wav b/r1-a/dataset/gsm8k_with_audio/test/1103.wav
new file mode 100644
index 0000000000000000000000000000000000000000..330df62ef52383a047c114e86d1033b944e50b50
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1103.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ca895ac09a5ca6b2534a2aa0561e9947a7aacd75789c050c2fa8c6eab8cc54d
+size 1578320
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1104.wav b/r1-a/dataset/gsm8k_with_audio/test/1104.wav
new file mode 100644
index 0000000000000000000000000000000000000000..64ddaaf7ac2162d070eb4de7b272eb2454844461
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1104.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84680bd666e6e3931946b5ddfb67d4c4c816dc5c21820c0dd66b0713cca1665b
+size 664400
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1105.wav b/r1-a/dataset/gsm8k_with_audio/test/1105.wav
new file mode 100644
index 0000000000000000000000000000000000000000..2802f417a192188b0758f5dfce0e31233bc9b0d6
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1105.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1a42cf18b4f578f9693f9cbbd7ef169a088e66c8fb32f9a46e4cdf59de7018e
+size 1267280
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1107.wav b/r1-a/dataset/gsm8k_with_audio/test/1107.wav
new file mode 100644
index 0000000000000000000000000000000000000000..9270b3ed2bea128a35d8a64d9cac049d851dd46f
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1107.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e6411751e4dce09e4621d0e739d66ba3ac444758d4f3b5044d950ef3c286d9b7
+size 675920
diff --git a/r1-a/dataset/gsm8k_with_audio/test/111.wav b/r1-a/dataset/gsm8k_with_audio/test/111.wav
new file mode 100644
index 0000000000000000000000000000000000000000..9ae92abac07d333cef6a3150c1f79cc13dbdab3b
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/111.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f2996957b7c8a37b8d86bfabc3f3770952d39981b9d79cd12f8537bdcbd189a
+size 1459280
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1112.wav b/r1-a/dataset/gsm8k_with_audio/test/1112.wav
new file mode 100644
index 0000000000000000000000000000000000000000..c8cdd5a8c7707e43465866e5c6f15549ea6621a6
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1112.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8234ba1f1e907b102ea007d9abf5b2e07e3794bea1f2d334686f971bfd647416
+size 1063760
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1113.wav b/r1-a/dataset/gsm8k_with_audio/test/1113.wav
new file mode 100644
index 0000000000000000000000000000000000000000..b7ba9c374e6bd53095c5e7207d4c3929330d250a
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1113.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:948a38997b2b668bf60e0cb5cf41be4a40fd871f5fff8d4fc29bf19c524397b0
+size 944720
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1117.wav b/r1-a/dataset/gsm8k_with_audio/test/1117.wav
new file mode 100644
index 0000000000000000000000000000000000000000..95a04dd4a50ce962ab4b8ebf81b669b87f8a6aac
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1117.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1e8fc77b9233cdc86dff0850c0a5f1c88458f37ffa39d45f8f020bc6aa8e79d
+size 837200
diff --git a/r1-a/dataset/gsm8k_with_audio/test/1121.wav b/r1-a/dataset/gsm8k_with_audio/test/1121.wav
new file mode 100644
index 0000000000000000000000000000000000000000..b43736f3f065408ad034a9fad7c54b6822e0b72e
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/1121.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:842195f45341fbaf7b8c769a96a24d7806aa228843634436ab2a65851da12a80
+size 960080
diff --git a/r1-a/dataset/gsm8k_with_audio/test/26.wav b/r1-a/dataset/gsm8k_with_audio/test/26.wav
new file mode 100644
index 0000000000000000000000000000000000000000..03f2b42f591071c1a7ba262c6d4b06eae496f750
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/26.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fbfcfdd59911a354fb11358b72d11817b50abbd5c7593691c6df8dc60594bd00
+size 1812560
diff --git a/r1-a/dataset/gsm8k_with_audio/test/291.wav b/r1-a/dataset/gsm8k_with_audio/test/291.wav
new file mode 100644
index 0000000000000000000000000000000000000000..b8b23950c15ce03718510946da9490cdfcdd9176
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/291.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61658ad3b1e39853faf5bc2fe887d84f1690197677854d4a0a69f2f94f5bac90
+size 1336400
diff --git a/r1-a/dataset/gsm8k_with_audio/test/297.wav b/r1-a/dataset/gsm8k_with_audio/test/297.wav
new file mode 100644
index 0000000000000000000000000000000000000000..32069795f119c503657ce49f2fc35c62e9f355a9
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/297.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:413028fcba9b4836c60bf1a6abde24d21407f464778db41d78f67e569a85b974
+size 1509200
diff --git a/r1-a/dataset/gsm8k_with_audio/test/3.wav b/r1-a/dataset/gsm8k_with_audio/test/3.wav
new file mode 100644
index 0000000000000000000000000000000000000000..3c0d8a12cd965324590079c8dcb548e3938b8941
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/3.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ecc76a295f21900e69caa7f818346c2a0c85345e55e99a8adbb778d771b7f4a
+size 722000
diff --git a/r1-a/dataset/gsm8k_with_audio/test/306.wav b/r1-a/dataset/gsm8k_with_audio/test/306.wav
new file mode 100644
index 0000000000000000000000000000000000000000..db1b1828ace7855d237fd63a023183831980df5f
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/306.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37fd95c90c092df493a2c598409a47ec93ce0b3f5f1d3d6e348b33e22c72b6a4
+size 1094480
diff --git a/r1-a/dataset/gsm8k_with_audio/test/307.wav b/r1-a/dataset/gsm8k_with_audio/test/307.wav
new file mode 100644
index 0000000000000000000000000000000000000000..c44cfd9213a1ae0b2103f0ab184c5dea961af64c
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/307.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a9b279d3797fc2c2fbd3bada6075cc75b38dd4b5d3e89c5bf83bae4b2861264
+size 1305680
diff --git a/r1-a/dataset/gsm8k_with_audio/test/308.wav b/r1-a/dataset/gsm8k_with_audio/test/308.wav
new file mode 100644
index 0000000000000000000000000000000000000000..0740965f1ce3c9485c761c4ab8273aaf2a38285e
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/308.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b75f799ceaae4bfd672d887ef7f64bb8bed4169cf0d528c634eade5861bf4eec
+size 1321040
diff --git a/r1-a/dataset/gsm8k_with_audio/test/310.wav b/r1-a/dataset/gsm8k_with_audio/test/310.wav
new file mode 100644
index 0000000000000000000000000000000000000000..3ce825546ad5e6fd3ceeaf236dfce00e506019e3
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/310.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b34bd5ace32887ba81b1b1875975ffa132bd0f2e5519a292dbee3a009e66c356
+size 1946960
diff --git a/r1-a/dataset/gsm8k_with_audio/test/311.wav b/r1-a/dataset/gsm8k_with_audio/test/311.wav
new file mode 100644
index 0000000000000000000000000000000000000000..2e741c9f5bc2a82d6d704359e14813a8270e5152
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/311.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0813093cb313dae0d37bb4bfade28df303d7d9a9752248a1a14980ed1abe977
+size 1267280
diff --git a/r1-a/dataset/gsm8k_with_audio/test/313.wav b/r1-a/dataset/gsm8k_with_audio/test/313.wav
new file mode 100644
index 0000000000000000000000000000000000000000..b3c34a5c32fc5255c1c955895f014e06ec303919
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/313.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:546e2d8053c063cf4c0a63aedb3c12bd782426d3d36bae88a5f21e28ae1c28ad
+size 1586000
diff --git a/r1-a/dataset/gsm8k_with_audio/test/34.wav b/r1-a/dataset/gsm8k_with_audio/test/34.wav
new file mode 100644
index 0000000000000000000000000000000000000000..8370f8c7c2aca77a4d718331425e68622974c4e8
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/34.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfb7e0f54d93ef4946e2ba74aeeb15b31347249083fb4dc732119e5a8cfb2f7b
+size 1236560
diff --git a/r1-a/dataset/gsm8k_with_audio/test/357.wav b/r1-a/dataset/gsm8k_with_audio/test/357.wav
new file mode 100644
index 0000000000000000000000000000000000000000..7a0a2a2c7e21e476a0070ad5336139170e09776a
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/357.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aab60d06342141bcbd67e2f7e99d7acf11a8037c683a96fc7355dfc29a0a5fa9
+size 1820240
diff --git a/r1-a/dataset/gsm8k_with_audio/test/361.wav b/r1-a/dataset/gsm8k_with_audio/test/361.wav
new file mode 100644
index 0000000000000000000000000000000000000000..ef8dfee55fed5c3f021467b66036053c5691405c
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/361.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04fba3aea3e60680a4ce65bb3346b7004c0278b633cf3057ba24032945c09b5e
+size 2254160
diff --git a/r1-a/dataset/gsm8k_with_audio/test/366.wav b/r1-a/dataset/gsm8k_with_audio/test/366.wav
new file mode 100644
index 0000000000000000000000000000000000000000..0a6f6971c2142baacffb1b715fdd35e0e15dc697
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/366.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76342cf72020b2d868e243e3c646624a73d74b36cd00384038aeb7c34d11fe50
+size 1532240
diff --git a/r1-a/dataset/gsm8k_with_audio/test/367.wav b/r1-a/dataset/gsm8k_with_audio/test/367.wav
new file mode 100644
index 0000000000000000000000000000000000000000..4d736336b6b34c591fff2428601bd3e97e3d568d
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/367.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:20f38b6381b99b090fa8e7cbe411745b47a2ff53f48f8629c0495273115172da
+size 2081360
diff --git a/r1-a/dataset/gsm8k_with_audio/test/368.wav b/r1-a/dataset/gsm8k_with_audio/test/368.wav
new file mode 100644
index 0000000000000000000000000000000000000000..8cf4d7dd32ec9266bd103c3b5965d1d057e69242
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/368.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09620064d6282031059e682a2941baf93ec25428a2d5582b6273419b08d8cbd1
+size 844880
diff --git a/r1-a/dataset/gsm8k_with_audio/test/37.wav b/r1-a/dataset/gsm8k_with_audio/test/37.wav
new file mode 100644
index 0000000000000000000000000000000000000000..b0dece0dc213aa1b6a2d13d154a3455fadeae9fc
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/37.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cad75f8356ae3baefd7a0ef8835e214372227d1ccef5e281a0d23d4a495e61f2
+size 1966160
diff --git a/r1-a/dataset/gsm8k_with_audio/test/370.wav b/r1-a/dataset/gsm8k_with_audio/test/370.wav
new file mode 100644
index 0000000000000000000000000000000000000000..f661995921448ad59c87d6b8d6ca48ee4f1db7ec
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/370.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1860044c383aa2aafb72fa50ee18b1dd538aa9c27647925587d1af53e50aeaa
+size 1228880
diff --git a/r1-a/dataset/gsm8k_with_audio/test/378.wav b/r1-a/dataset/gsm8k_with_audio/test/378.wav
new file mode 100644
index 0000000000000000000000000000000000000000..e407ec1946249aabb175c8f840d3af5a9fa63f15
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/378.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ea8f37bb77225aab0700f0b93d6faec054b8f8e2a6e860e4d18b05a53ffdc2f
+size 1109840
diff --git a/r1-a/dataset/gsm8k_with_audio/test/379.wav b/r1-a/dataset/gsm8k_with_audio/test/379.wav
new file mode 100644
index 0000000000000000000000000000000000000000..d66259a57b608a1d8253a06f0161f7d480875025
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/379.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19625e3cf3c1bd98825f6be5479cf81670185a71bdad8e3d1eab4553052f11c1
+size 1394000
diff --git a/r1-a/dataset/gsm8k_with_audio/test/38.wav b/r1-a/dataset/gsm8k_with_audio/test/38.wav
new file mode 100644
index 0000000000000000000000000000000000000000..f475e3031508d544d35369a190eccecb4dca06cf
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/38.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7dd32f9e447bc3961a063e2b6d1e6a87533fb75b19a7ba7d2eec9c3a52fec10
+size 1248080
diff --git a/r1-a/dataset/gsm8k_with_audio/test/382.wav b/r1-a/dataset/gsm8k_with_audio/test/382.wav
new file mode 100644
index 0000000000000000000000000000000000000000..4175e66786e5e9b9c3198a862eda56933b4a4b36
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/382.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6416f01c53846a2c071bf1d3cf89549d6cbd4c4a23daf6f70286aa6291669a3
+size 1221200
diff --git a/r1-a/dataset/gsm8k_with_audio/test/386.wav b/r1-a/dataset/gsm8k_with_audio/test/386.wav
new file mode 100644
index 0000000000000000000000000000000000000000..ec77bbcc5dc96fd7b507d7cf18804fe63cc3a731
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/386.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7d69222c42ed3bba5aa8f386427812b2d9b0d105bcc22e29ffb0a7f5dd8cded
+size 2039120
diff --git a/r1-a/dataset/gsm8k_with_audio/test/387.wav b/r1-a/dataset/gsm8k_with_audio/test/387.wav
new file mode 100644
index 0000000000000000000000000000000000000000..60aa327f9422e7bfea9dc1754c88e07555036f18
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/387.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de36666a2d1d113d54a36396a674ed664729732720d0367af11fde119c8454ad
+size 714320
diff --git a/r1-a/dataset/gsm8k_with_audio/test/390.wav b/r1-a/dataset/gsm8k_with_audio/test/390.wav
new file mode 100644
index 0000000000000000000000000000000000000000..265fade7a456d369fdf53f385c29afee29d5bc4c
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/390.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8c1510387384d9f53b0b7595bc22d73cfc2fa413bab4d7b6e241d4fa7ee4300b
+size 1524560
diff --git a/r1-a/dataset/gsm8k_with_audio/test/396.wav b/r1-a/dataset/gsm8k_with_audio/test/396.wav
new file mode 100644
index 0000000000000000000000000000000000000000..54e20ba112a7e04ca87d100999b817b2b0899bd8
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/396.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c14d3403ed515415f90117eab1039ac41cde94dcfdde05d13d25af592f33aa6
+size 1378640
diff --git a/r1-a/dataset/gsm8k_with_audio/test/398.wav b/r1-a/dataset/gsm8k_with_audio/test/398.wav
new file mode 100644
index 0000000000000000000000000000000000000000..c49cefe3ebae4036df6912383bac028ae74db752
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/398.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a955f2661e623117c82ba26f52c4fad93def6c4cc98dc4a9ee06b94a60594489
+size 1470800
diff --git a/r1-a/dataset/gsm8k_with_audio/test/399.wav b/r1-a/dataset/gsm8k_with_audio/test/399.wav
new file mode 100644
index 0000000000000000000000000000000000000000..2d10c520c543f79c3f522280b0ffae234055d022
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/399.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa61fa30c3b17cc658856675f323e19068ae33d99d9bb0576acfd8a366624e9c
+size 1117520
diff --git a/r1-a/dataset/gsm8k_with_audio/test/441.wav b/r1-a/dataset/gsm8k_with_audio/test/441.wav
new file mode 100644
index 0000000000000000000000000000000000000000..51f2eca8f6615e1a9a3b92c0397c8eb95ac5d178
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/441.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6aea3fd598cc6a2d10cd5ba4c1109304d34c4eb59365961146315e4c4966bf77
+size 1409360
diff --git a/r1-a/dataset/gsm8k_with_audio/test/442.wav b/r1-a/dataset/gsm8k_with_audio/test/442.wav
new file mode 100644
index 0000000000000000000000000000000000000000..e9de7a94a82fcd4f903532d984ecd133d6f3cd8b
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/442.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0eaba4ae664275880b46a74ffd24b80dd48c7262dcd7bd6036e8d93327d644b0
+size 1728080
diff --git a/r1-a/dataset/gsm8k_with_audio/test/444.wav b/r1-a/dataset/gsm8k_with_audio/test/444.wav
new file mode 100644
index 0000000000000000000000000000000000000000..7ee96eda1ed3553590d6848f6c252fd182651d4e
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/444.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f3f5226514b6682acbfe9aaf302042f0c278b28ad9ff5345a1d3ffc2fe44801
+size 1079120
diff --git a/r1-a/dataset/gsm8k_with_audio/test/445.wav b/r1-a/dataset/gsm8k_with_audio/test/445.wav
new file mode 100644
index 0000000000000000000000000000000000000000..a97dea44fd9bea92d1f3ee1c88d5655fde183821
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/445.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:025148de0657627901a48c16c4f36c68a182c2c13cdda4b3f4b1dfe580624b59
+size 1762640
diff --git a/r1-a/dataset/gsm8k_with_audio/test/449.wav b/r1-a/dataset/gsm8k_with_audio/test/449.wav
new file mode 100644
index 0000000000000000000000000000000000000000..9c1696efd31548c6a79fd546ba78c68add018754
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/449.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0df03845c3a7a3822aa2cb8b58cff3403f5c3b4faa81b05e0bd7222d7e926fc
+size 1413200
diff --git a/r1-a/dataset/gsm8k_with_audio/test/451.wav b/r1-a/dataset/gsm8k_with_audio/test/451.wav
new file mode 100644
index 0000000000000000000000000000000000000000..3c4fc8b81f95237aa2f4c3f690a013d8fd24a5d6
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/451.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7cec4e3fee27b8e8dd8641e8a100d63f921887bedef19f26d157252e19781c7b
+size 2000720
diff --git a/r1-a/dataset/gsm8k_with_audio/test/453.wav b/r1-a/dataset/gsm8k_with_audio/test/453.wav
new file mode 100644
index 0000000000000000000000000000000000000000..5588d0aac7fec478fd1671f3f660603a23f229ea
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/453.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1635c5324b81e86a80f8d15c9c4eae0c9e5ca9648048a536fd6c9159ae0f530
+size 1282640
diff --git a/r1-a/dataset/gsm8k_with_audio/test/455.wav b/r1-a/dataset/gsm8k_with_audio/test/455.wav
new file mode 100644
index 0000000000000000000000000000000000000000..598bdb872fc7ce2139f71626d8b67544bf655581
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/455.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:685d2a471fc23ccca651f028cda3b2e4569d66fded6468bd649e20a90771d017
+size 1705040
diff --git a/r1-a/dataset/gsm8k_with_audio/test/456.wav b/r1-a/dataset/gsm8k_with_audio/test/456.wav
new file mode 100644
index 0000000000000000000000000000000000000000..924ba21d17a49e84e87ebf52e5109aa81559625b
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/456.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5068284acab37341d70b47664e3cd206e3ed85c0b782bd5549b97e70eea62ac8
+size 1033040
diff --git a/r1-a/dataset/gsm8k_with_audio/test/462.wav b/r1-a/dataset/gsm8k_with_audio/test/462.wav
new file mode 100644
index 0000000000000000000000000000000000000000..99ba6089b0516b20131e4174d9d71eb2cf163d34
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/462.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81431337b1acb42f893504318534aed7813d8cddc71ffa4f10b96ceedfe7e8fa
+size 745040
diff --git a/r1-a/dataset/gsm8k_with_audio/test/506.wav b/r1-a/dataset/gsm8k_with_audio/test/506.wav
new file mode 100644
index 0000000000000000000000000000000000000000..ec45a9f0c2600b40997bf5af0e31ceced52176f4
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/506.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e80bbc6d5056898e372e0997cc568fb6bc8ddad6ca8ecd0944e5ee330d65e49
+size 1298000
diff --git a/r1-a/dataset/gsm8k_with_audio/test/520.wav b/r1-a/dataset/gsm8k_with_audio/test/520.wav
new file mode 100644
index 0000000000000000000000000000000000000000..865213a24b91f9a35b14e7f0081c716e9563a437
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/520.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c4d50b1e44276c532b8f685b4304b6962a048b364bb4c181591ee1477ae7fda
+size 2200400
diff --git a/r1-a/dataset/gsm8k_with_audio/test/53.wav b/r1-a/dataset/gsm8k_with_audio/test/53.wav
new file mode 100644
index 0000000000000000000000000000000000000000..2b3e8984f608849f1e32e981900e008d98396771
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/53.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f56ae13c67ca7ac4626145cfa29afd22acbcf47bace8fcbba419939f93c642e
+size 2330960
diff --git a/r1-a/dataset/gsm8k_with_audio/test/54.wav b/r1-a/dataset/gsm8k_with_audio/test/54.wav
new file mode 100644
index 0000000000000000000000000000000000000000..7fac4cbeac7c1b68fb3c131c52f7a91612c298db
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/54.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:076a58ed674405eec35cbb2bfd46cc4a1994240d223c40bd39c52d186c902bf4
+size 1989200
diff --git a/r1-a/dataset/gsm8k_with_audio/test/540.wav b/r1-a/dataset/gsm8k_with_audio/test/540.wav
new file mode 100644
index 0000000000000000000000000000000000000000..2d9c72b6b971c398a3114595f06f92f8cda33f90
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/540.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7367ac9044529f85fde96444d8df72fd38ffe64c2cac41e3823d0841913da366
+size 1639760
diff --git a/r1-a/dataset/gsm8k_with_audio/test/541.wav b/r1-a/dataset/gsm8k_with_audio/test/541.wav
new file mode 100644
index 0000000000000000000000000000000000000000..1fec9561b282f69f3a88c8010f8496b177f81452
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/541.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3254b2d44634e768147e9d34f56332d1de92e453aed47e16512af3e200520181
+size 1497680
diff --git a/r1-a/dataset/gsm8k_with_audio/test/542.wav b/r1-a/dataset/gsm8k_with_audio/test/542.wav
new file mode 100644
index 0000000000000000000000000000000000000000..b5ea516430ed21b3c793aa84d2d16bed71b489f1
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/542.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4082ab0f72004be8f64b554bd3dc0c94f0417a553f3453516b048f29d76bcc4b
+size 1417040
diff --git a/r1-a/dataset/gsm8k_with_audio/test/543.wav b/r1-a/dataset/gsm8k_with_audio/test/543.wav
new file mode 100644
index 0000000000000000000000000000000000000000..c09ed6e01f4ba4cba9c83d642f5798d84777a5f3
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/543.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2baafcd09643259fb659ada671d8b08966ac060e60ea3b0ffc8c0bf6e27598cb
+size 1040720
diff --git a/r1-a/dataset/gsm8k_with_audio/test/544.wav b/r1-a/dataset/gsm8k_with_audio/test/544.wav
new file mode 100644
index 0000000000000000000000000000000000000000..e5adbdc572a40c794b745599ccfa403b486e0cb9
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/544.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8cf80078fb2dd187331d5bb94399931460abb6465bff551f2678a57d9c5431d
+size 975440
diff --git a/r1-a/dataset/gsm8k_with_audio/test/545.wav b/r1-a/dataset/gsm8k_with_audio/test/545.wav
new file mode 100644
index 0000000000000000000000000000000000000000..3360dfb8bd62fae03b23ffb428972b24241d1a7f
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/545.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf88fa2f163c02924577ccf0a22fcac297941cdb57694c26ffdf59ce71794cd0
+size 960080
diff --git a/r1-a/dataset/gsm8k_with_audio/test/546.wav b/r1-a/dataset/gsm8k_with_audio/test/546.wav
new file mode 100644
index 0000000000000000000000000000000000000000..3c28c72cc46c94c14d67970f742dd2841fb14460
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/546.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8fbac0749d69fb89413cb5128f37f276c3bfd92a3d25fb16b301bf8d8a846cb2
+size 1655120
diff --git a/r1-a/dataset/gsm8k_with_audio/test/547.wav b/r1-a/dataset/gsm8k_with_audio/test/547.wav
new file mode 100644
index 0000000000000000000000000000000000000000..627a241b6c6ba2cc7178f2714f6bf17230d858a2
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/547.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b0b0e785b70e59dcfdc097024a5475e5bec765bb47ae6c2f740c233e1dea220
+size 1271120
diff --git a/r1-a/dataset/gsm8k_with_audio/test/548.wav b/r1-a/dataset/gsm8k_with_audio/test/548.wav
new file mode 100644
index 0000000000000000000000000000000000000000..1a08df869687201d8df3eae7013aff22c570f57b
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/548.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:daad7afb15a65a029c9619644f94553131011d76cceb7481c5ce085452892291
+size 718160
diff --git a/r1-a/dataset/gsm8k_with_audio/test/549.wav b/r1-a/dataset/gsm8k_with_audio/test/549.wav
new file mode 100644
index 0000000000000000000000000000000000000000..0aee4ff5fad52760fb7505242e025de9fe23e18c
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/549.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1f61c1e22d5e18c4b3579cb5b4a438e6d2e3f0ab5a6c9fa38cbc8a2785887fd
+size 1363280
diff --git a/r1-a/dataset/gsm8k_with_audio/test/551.wav b/r1-a/dataset/gsm8k_with_audio/test/551.wav
new file mode 100644
index 0000000000000000000000000000000000000000..09e7604c72e52418a2d22249d228dafbaa0d319d
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/551.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6fb055e76009c3469ccab435e75628237f8b65fd37fe58738836fe415eb5d3c2
+size 410960
diff --git a/r1-a/dataset/gsm8k_with_audio/test/552.wav b/r1-a/dataset/gsm8k_with_audio/test/552.wav
new file mode 100644
index 0000000000000000000000000000000000000000..d4904010f46e8e8bf5bda229db3fa2f6196777b5
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/552.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dcd21c0fb2ac0e3d10ee9d17ec11154829c30b71ceb092a0c11327699b38126a
+size 1774160
diff --git a/r1-a/dataset/gsm8k_with_audio/test/557.wav b/r1-a/dataset/gsm8k_with_audio/test/557.wav
new file mode 100644
index 0000000000000000000000000000000000000000..bfaed8f6ea90bb800cd381a8f69e801e5bdee8ca
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/557.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca6b516fe17ce9140680031aca7457c90db807629e749d480a520e3ddc97bed4
+size 983120
diff --git a/r1-a/dataset/gsm8k_with_audio/test/583.wav b/r1-a/dataset/gsm8k_with_audio/test/583.wav
new file mode 100644
index 0000000000000000000000000000000000000000..478beab0f1bd61a2131f4c35f58775412e35f272
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/583.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8efe699969877f9c9774213735d4e83b3ec30e50f3ae97a3f2c37a74425cbad
+size 1551440
diff --git a/r1-a/dataset/gsm8k_with_audio/test/602.wav b/r1-a/dataset/gsm8k_with_audio/test/602.wav
new file mode 100644
index 0000000000000000000000000000000000000000..970778c7a78d30b1b9aa687d6a9dfeb51cbee8ce
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/602.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d56a97af4b495d7fca91fff2160dbf031a5dddc12ab95a3c69d88cda9f0174e0
+size 979280
diff --git a/r1-a/dataset/gsm8k_with_audio/test/606.wav b/r1-a/dataset/gsm8k_with_audio/test/606.wav
new file mode 100644
index 0000000000000000000000000000000000000000..09f4ab21bbd06b10a6094af9b96fd5a1f64aeff9
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/606.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7be7a4da0379ffa7d19e25110e115d47450aad8411a32efbc38e53902966c198
+size 1459280
diff --git a/r1-a/dataset/gsm8k_with_audio/test/607.wav b/r1-a/dataset/gsm8k_with_audio/test/607.wav
new file mode 100644
index 0000000000000000000000000000000000000000..0b2f2580b3c34e5ca6d419f12f348b5469acc96d
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/607.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8bcc849ad8bde76cf334978e848025436b93b44a6ff38ff0e65bac5df1537281
+size 1336400
diff --git a/r1-a/dataset/gsm8k_with_audio/test/608.wav b/r1-a/dataset/gsm8k_with_audio/test/608.wav
new file mode 100644
index 0000000000000000000000000000000000000000..86f95237882bfd02fb8702244b2777132d64ec8f
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/608.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:321a781184d6a1d94d9cd98cf24fc98cd0061e3898975f06f826871639237ed7
+size 1897040
diff --git a/r1-a/dataset/gsm8k_with_audio/test/609.wav b/r1-a/dataset/gsm8k_with_audio/test/609.wav
new file mode 100644
index 0000000000000000000000000000000000000000..f6ea4113b8f183204e10fcab7392240eb93a44a2
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/609.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab440bf1045491ca809d0284689189b06a5cbab25dcb885430d282bc68f2f096
+size 1382480
diff --git a/r1-a/dataset/gsm8k_with_audio/test/61.wav b/r1-a/dataset/gsm8k_with_audio/test/61.wav
new file mode 100644
index 0000000000000000000000000000000000000000..c405fdf746773930a88e47cdc450349b90eade0f
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/61.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b23b42c6ff63b9158a8b20d543081ed557302143ed16bdbf46243c0bbaa43848
+size 1574480
diff --git a/r1-a/dataset/gsm8k_with_audio/test/610.wav b/r1-a/dataset/gsm8k_with_audio/test/610.wav
new file mode 100644
index 0000000000000000000000000000000000000000..3a8f7f1e35446f971c5360f7b856dcef86c6e420
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/610.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5928ad12a73572d118a31e8a859f40336c0029d5d2278094a2a33ffb0273c44b
+size 1225040
diff --git a/r1-a/dataset/gsm8k_with_audio/test/611.wav b/r1-a/dataset/gsm8k_with_audio/test/611.wav
new file mode 100644
index 0000000000000000000000000000000000000000..2a5fbb2d5ddc8aed61baa60230c298d3ec7ae34e
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/611.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e5de2c4175df98fd0a23361e02de57f3165cc1c040facba155b31859133e2b61
+size 1770320
diff --git a/r1-a/dataset/gsm8k_with_audio/test/613.wav b/r1-a/dataset/gsm8k_with_audio/test/613.wav
new file mode 100644
index 0000000000000000000000000000000000000000..f1e177dc083768015c63256008c83b8b6b1a60e6
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/613.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ae73522e029abde2686a370d3301b6f482c94492c95e2e0acb22b37f581a0ec
+size 887120
diff --git a/r1-a/dataset/gsm8k_with_audio/test/614.wav b/r1-a/dataset/gsm8k_with_audio/test/614.wav
new file mode 100644
index 0000000000000000000000000000000000000000..20748f13ca7ad0ad5a243e58d421d9ec5dab8f67
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/614.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b5330a2a28b69d198a2db482f5410e111d85d03b0d7f6d56e4d9c9e7a2a7bb3
+size 1363280
diff --git a/r1-a/dataset/gsm8k_with_audio/test/616.wav b/r1-a/dataset/gsm8k_with_audio/test/616.wav
new file mode 100644
index 0000000000000000000000000000000000000000..d9aa0667f689125df4de219853d650577749a7b4
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/616.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b5753302877afa50d92740cf06c1587cf8a3b0a14fc6daeb2740c3552e2d6a4
+size 1309520
diff --git a/r1-a/dataset/gsm8k_with_audio/test/617.wav b/r1-a/dataset/gsm8k_with_audio/test/617.wav
new file mode 100644
index 0000000000000000000000000000000000000000..3fcf99dfe2151852dd811e5ff6dfdf33e7360e09
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/617.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d44f991328fd803331a38adfb686d7d48c92a646e804663bd76e9686ed0fa1d5
+size 460880
diff --git a/r1-a/dataset/gsm8k_with_audio/test/618.wav b/r1-a/dataset/gsm8k_with_audio/test/618.wav
new file mode 100644
index 0000000000000000000000000000000000000000..d67e6ec912469d1d1dd3c2278fd2543e36614d82
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/618.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48ff93eb7d7e703dd1333761f6a1cee4b07eb29e388b6178512324731621f01d
+size 649040
diff --git a/r1-a/dataset/gsm8k_with_audio/test/619.wav b/r1-a/dataset/gsm8k_with_audio/test/619.wav
new file mode 100644
index 0000000000000000000000000000000000000000..71d95a4392a19a53117709f023f4d9cc228d87ad
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/619.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2ce77e39cb3ecb7eb488559dc7cf327acd7590a2e3fc747161ffc13634f8cb5
+size 1493840
diff --git a/r1-a/dataset/gsm8k_with_audio/test/62.wav b/r1-a/dataset/gsm8k_with_audio/test/62.wav
new file mode 100644
index 0000000000000000000000000000000000000000..7040496b18a2320e6927463cbe23fa7f0b160552
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/62.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a68009903102e585dfb809224fa06524e9b9c0f40a7b3301228c6cd87563c27d
+size 1386320
diff --git a/r1-a/dataset/gsm8k_with_audio/test/620.wav b/r1-a/dataset/gsm8k_with_audio/test/620.wav
new file mode 100644
index 0000000000000000000000000000000000000000..170bfd302c0130d122f1b1fce47e1985b935b823
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/620.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:92259b87ea2b66a19d37d695e2bf24454d8b25e52d28582124e722c26131e25b
+size 1313360
diff --git a/r1-a/dataset/gsm8k_with_audio/test/623.wav b/r1-a/dataset/gsm8k_with_audio/test/623.wav
new file mode 100644
index 0000000000000000000000000000000000000000..5ca8c85e47add7e2d30a65db652ca7c6a5d7110b
--- /dev/null
+++ b/r1-a/dataset/gsm8k_with_audio/test/623.wav
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:031e5179d8a9efb72cbe1e7c238725ea70a3d38cedd82a2b711f2d8815f004b6
+size 610640