text stringlengths 123 13.3k | audio audioduration (s) 9.77 277 ⌀ | modality stringclasses 2
values | category class label 2
classes | source_id stringclasses 3
values | source_name stringclasses 3
values | source_license stringclasses 2
values | source_url stringclasses 3
values | original_id stringlengths 11 24 | pii_redacted bool 2
classes | pii_filtered bool 1
class |
|---|---|---|---|---|---|---|---|---|---|---|
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_3865 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_652 | false | true | |
Innocent: Hello. Suspect: Hello! My name is Karen and I'm calling from Speedy Deliveries. I have a delivery scheduled for you today and I just wanted to confirm a few details to make sure everything goes smoothly. Can you please confirm your address is 145 Oak Street, Apartment 3B? Innocent: Yes, that's correct. But, ... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_test_47 | true | true |
Innocent: Hello. Suspect: Hi there! My name is Rachel and I'm an insurance sales representative. How are you today? Innocent: Hi Rachel, I'm doing great, thanks! How can I help you? Suspect: I'm calling today to discuss your insurance options. We're offering some new policies that might be of interest to you. Can you ... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_test_118 | false | true |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_1900 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_3123 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_2018 | false | true | |
Innocent: Hello. Suspect: Hello, this is Officer Rachel from the Social Security Administration. How are you today, [NAME]? We've been trying to reach you regarding a very serious matter. Your social security number has been suspended due to suspicious activity. I need to verify some information to resolve this issue.... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_train_942 | true | true |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_3195 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_38 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_2929 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_2011 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_3426 | false | true | |
Innocent: Hello. Suspect: Hey Sarah, how's it going? I was just calling to finalize the plans for our trip to the beach this weekend. Did you end up booking that hotel room like we discussed? Innocent: Who's Sarah? I think you must have the wrong number. I'm not Sarah. Who is this? Suspect: I'm so sorry! I must have d... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_train_1176 | false | true |
Innocent: Hello. Suspect: Hello, this is Alex from TechGuard Solutions. We've been monitoring your computer's activity and have detected a severe virus infection. Your device is at risk of crashing and losing all your important files. Can you please tell me what kind of antivirus software you're currently using? Innoc... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_train_1041 | false | true |
Innocent: Hello. Suspect: Hello MJL1v, my name is Officer Rachel Thompson from the Social Security Administration. I'm calling to inform you that we've received a suspicious activity report on your social security number. It appears that someone has been using your number to commit fraud. I need to verify some informa... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_test_239 | true | true |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_3649 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_2012 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_176 | false | true | |
Innocent: Hello. Suspect: Hello! My name is Alex and I'm calling from TechShield Solutions. We've been monitoring your internet activity and it appears that your computer has been infected with a malicious virus. It's a very serious issue and if not addressed immediately, it could lead to data loss and identity theft.... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_test_275 | false | true |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_test_285 | false | true | |
Innocent: Hello. Suspect: Hey Sarah! How's it going? I was thinking we could grab lunch today to discuss the Johnson account. Innocent: I'm not Sarah, and I don't know what you're talking about. Who is this? Suspect: I'm so sorry! I must have dialed the wrong number. I'm Alex, and I was trying to reach my colleague Sa... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_train_1256 | false | true |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_2614 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_1659 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_2139 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_1390 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_test_270 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_3283 | false | true | |
Innocent: Hello. Suspect: Hello, is this [NAME]? I'm calling from the Consumer Protection Agency. We've been trying to reach you about a prize you've won in our national sweepstakes. You've been selected to receive a cash award of $10,000! Can you please confirm your mailing address so we can process the paperwork? In... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_test_173 | true | true |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_1623 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_726 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_3694 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_2477 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_379 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_3542 | false | true | |
Hi Mom, just calling to check in. How was your doctor's appointment? Dad and I were thinking of coming to visit next weekend if that works for you. | Not supported with pagination yet | text | 0not_scam | synthetic-elder-scam | Claude-generated elder-scam scenarios | CC-BY-4.0 | synthetic_14 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_1064 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_test_346 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_3962 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_2880 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_993 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_2967 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_6 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_test_63 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_3589 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_3107 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_3023 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_2333 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_test_231 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_1469 | false | true | |
Innocent: Hello. Suspect: Hello, [NAME]. This is Officer Thompson from the Social Security Administration. We've been trying to reach you about a serious issue with your social security number. It's been flagged for fraudulent activity and we need to verify some information to resolve this matter. Can you please confi... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_train_831 | true | true |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_373 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_3135 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_2732 | false | true | |
Innocent: Hello. Suspect: Hello, this is Alex from TechShield Solutions. We've been monitoring your computer's activity and we've detected some suspicious malware infections. Our system shows that your device is at high risk of data breach and compromise. I'm calling to help you resolve this issue before it's too late... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_train_1098 | false | true |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_945 | false | true | |
Innocent: Hello. Suspect: Hi, this is Karen from MedPro Health Services. I'm calling to confirm your appointment with [NAME] on Thursday at 2 PM. Can you please confirm that you're still able to make it? Innocent: Yeah, I'm still good for Thursday. But before we confirm, I was wondering if you guys offer any kind of r... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_train_96 | true | true |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_test_23 | false | true | |
Innocent: Hello. Suspect: Hi there! My name is Rachel and I'm an insurance sales representative. How are you today? Innocent: I'm doing great, Rachel. What can you do for me? I'm always open to exploring ways to save some money or make a smart investment. Suspect: I'm glad to hear that. I specialize in helping people ... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_train_417 | false | true |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_3317 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_3989 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_1401 | false | true | |
Innocent: Hello. Suspect: Hello! My name is Karen and I'm calling from Speedy Deliveries. I'm calling to confirm the details of your delivery scheduled for today. Can you please confirm your name and the address where you'd like the package delivered? Innocent: Good morning Karen. My name is Emily Wilson, and I'm expe... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_train_266 | false | true |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_3866 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_3878 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_1713 | false | true | |
Innocent: Hello. Suspect: Hi, this is Karen from Elite Services. I'm calling to confirm your appointment with [NAME] on Friday at 2 PM. Can you confirm that you're still able to make it? Innocent: Yes, Karen, that's correct. I'm still on for Friday at 2 PM with [NAME]. Thank you for calling to confirm. Suspect: You're... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_train_116 | true | true |
Innocent: Hello. Suspect: Hello! My name is Karen and I'm calling from Speedy Deliveries. I'm calling to confirm the details of your delivery scheduled for today. Can you please confirm your name and address for me? Innocent: Karen? Speedy Deliveries? I don't remember ordering anything. What's being delivered again? S... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_test_53 | true | true |
Innocent: Hello. Suspect: Hi there! My name is Rachel and I'm an insurance representative. How are you today? Innocent: Cut the small talk, Rachel. What do you want? Suspect: I apologize if I came on too strong. I'm calling today to discuss your insurance options. Are you currently insured, and if so, are you happy wi... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_train_336 | false | true |
Innocent: Hello. Suspect: Hi, this is Karen from MedCare Services. I'm calling to confirm your upcoming appointment with [NAME] on Thursday at 2 PM. Can you confirm that you're still able to make it? Innocent: Yes, I think so. I hope so. I really need to see the doctor. Is everything okay? Is there a problem with the ... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_train_20 | true | true |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_62 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_2942 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_1782 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_2595 | false | true | |
Innocent: Hello. Suspect: Hi, this is Karen from Smith Medical Group. I'm calling to confirm your appointment with [NAME] on Thursday at 2 PM. Can you confirm that you're still able to make it? Innocent: Yeah, that's correct. 2 PM Thursday. I'll be there. Suspect: Great, thank you for confirming, [NAME]. We're looking... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_train_60 | true | true |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_2511 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_1996 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_1285 | false | true | |
Innocent: Hello. Suspect: Hi, this is Karen from MedPro Health Services. I'm calling to confirm your upcoming appointment with [NAME] on Thursday at 2 PM. Can you please confirm that you'll be able to make it? Innocent: Oh, yes, I think so. I mean, I hope so. I don't want to miss it, but what if something comes up? Ca... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_train_22 | true | true |
Innocent: Hello. Suspect: Hi, this is Karen from Smith's Auto Repair. I'm calling to confirm your appointment for a routine oil change on your Toyota Camry, scheduled for tomorrow at 2 PM. Can you please confirm that you're still available to bring your car in at that time? Innocent: What's the point of calling me to ... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_train_7 | true | true |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_1858 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_2479 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_665 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_3065 | false | true | |
Innocent: Hello. Suspect: Hi there! My name is Rachel and I'm an insurance sales representative. How are you today? Innocent: I'm doing great, thanks! Nice to meet you, Rachel. Suspect: Nice to meet you too! I'm calling today because we've been expanding our services in your area and I wanted to reach out to see if yo... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_train_475 | false | true |
Innocent: Hello. Suspect: Hello, is this Emily Wilson? My name is Rachel Thompson and I'm calling from National Billing Services. How are you today? Innocent: Yes, that's me. I'm doing well, thank you. What can I do for you, Rachel? Suspect: Emily, I'm calling to inform you that our system has detected an overpayment ... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_train_620 | true | true |
Innocent: Hello. Suspect: Hello, [NAME]. My name is Officer James from the Social Security Administration. We've been trying to reach you about a serious issue with your social security number. It's been flagged for fraudulent activity and we need to verify some information to resolve this matter. Can you please confi... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_train_803 | true | true |
Innocent: Hello. Suspect: Hi there! My name is Alex and I'm calling from XYZ Insurance. How are you today? Innocent: Hello Alex, I'm doing well, thank you for asking. What can I do for you today? Suspect: I'm glad to hear that! I'm calling to discuss your insurance options. We offer a range of products, including life... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_test_106 | false | true |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_155 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_test_288 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_3760 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_454 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_181 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_3035 | false | true | |
Innocent: Hello. Suspect: Hi there! My name is Rachel and I'm an insurance sales representative. How are you doing today? Innocent: I'm doing great, Rachel. What kind of insurance are you selling? Is it something that can help me save some money or get a good return on investment? Suspect: I'm glad you're doing great!... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_train_410 | false | true |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_1764 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_1656 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 0not_scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_train_1173 | false | true | |
根据听到的音频内容，分析该通话是否涉及诈骗。
请按照以下格式输出:
"<thinking>
...(分析过程)
</thinking>
<answer>
{
"reason": "<reason_for_judgment>",
"confidence": <confidence_level>,
"is_fraud": <true/false>
}
</answer>"
说明：
- `reason`：字符串，简要总结判断原因。
- `confidence`：浮点数，表示对判断结果的置信度，范围从0到1，1表示完全置信。
- `is_fraud`：布尔值，表示该段音频是否涉及诈骗。`true`表示涉诈，`false`表示... | text+audio | 1scam | teleantifraud | TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | teleantifraud_test_117 | false | true | |
Innocent: Hello. Suspect: Hello, this is Alex from TechShield Solutions. We've been monitoring your computer's activity and have detected some suspicious malware infections. Our system shows that your device has been compromised and is at risk of data breach. Can you please tell me what kind of computer you're using a... | Not supported with pagination yet | text | 1scam | bothbosu-multi-agent | BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | bothbosu_train_1078 | false | true |
End of preview. Expand in Data Studio
GrandgemMa-Corpus (ScamBench)
A CC-BY 4.0 licensed multimodal scam-classification corpus for training and evaluating on-device scam detectors targeting elder safety.
Supports both text-only and multimodal (text+audio) fine-tuning. TeleAntiFraud rows include embedded audio (mp3), enabling Gemma 4 to learn from prosodic cues (urgency, pressure, tone) that transcription destroys.
Column Schema
| Column | Type | Description |
|---|---|---|
| text | string | Text content (PII filtered) |
| audio | Audio | Embedded audio (null for text-only rows; mp3 bytes stored in Parquet) |
| modality | string | "text" or "text+audio" |
| category | ClassLabel | 0=not_scam, 1=scam |
| source_id | string | Machine-readable source identifier |
| source_name | string | Human-readable source name |
| source_license | string | License of the source dataset |
| source_url | string | URL to the source dataset |
| original_id | string | Original row ID in the source dataset |
| pii_redacted | bool | Whether PII was found and redacted |
| pii_filtered | bool | Whether the PII filter was applied |
Using Audio Rows
from datasets import load_dataset
ds = load_dataset("s23deepak/GrandgemMa-Corpus")
for row in ds["train"]:
if row["audio"] is not None:
waveform = row["audio"]["array"] # numpy array
sr = row["audio"]["sampling_rate"] # 16000
# Feed waveform + text to multimodal model
Source Attribution
| Source | License | URL | Row Count |
|---|---|---|---|
| BothBosu/multi-agent-scam-conversation | Apache-2.0 | https://huggingface.co/datasets/BothBosu/multi-agent-scam-conversation | 1600 |
| Claude-generated elder-scam scenarios | CC-BY-4.0 | 19 | |
| TeleAntiFraud (multimodal) | Apache-2.0 | https://huggingface.co/datasets/JimmyMa99/TeleAntiFraud | 4400 |
Intended Use
Training and evaluating on-device scam classifiers for elder safety. Part of the GrandgemMa hackathon submission (Gemma 4 Good, 2026-05-17).
- Downloads last month
- 115