callID stringlengths 11 21 | recordID stringlengths 8 8 | source stringclasses 3
values | process stringclasses 1
value | system_prompt stringclasses 1
value | input stringlengths 1.47k 6.17k | thinking stringlengths 707 3.43k | output stringlengths 1.03k 3.25k | messages listlengths 4 4 | generated_instruction stringlengths 100 388 |
|---|---|---|---|---|---|---|---|---|---|
C-117173030 | 70686119 | RTN | xulytuchoi | You are a QC AI scoring the XỬ LÝ TỪ CHỐI skill of telesales staff.
Your only job: read transcript → score 9 steps → output JSON only.
DO NOT add any text outside JSON.
CORE PRINCIPLE
Evaluate TELESALE ACTION only — not customer response.
Telesale acted correctly → S regardless of customer outcome.
ACTION > OUTCOME.
... | Module: xulytuchoi Transcript section: [customer] [01:56 - 01:59] Anh chưa mua đâu nhé. Anh chưa mua đâu.
[telesales] [02:00 - 02:04] À dạ. Anh vậy không biết là anh có À chưa có nhu cầu ha.
[telesales] [02:04 - 02:10] Dạ thì em biết là hồi trước giờ là như anh chia sẻ hồi nãy á thì em cũng hiểu là cũng lâu rồi anh cũn... | I need to analyze this transcript for the XỬ LÝ TỪ CHỐI (objection handling) skill across 3 rounds of objections.
**Identifying the rounds:**
**Round 1:**
- Customer objection [01:56]: "Anh chưa mua đâu nhé. Anh chưa mua đâu."
- Step 1 (Làm dịu) [02:04-02:10]: Telesale shows empathy - "em cũng hiểu là cũng lâu rồi an... | [
{"id":1,"r":"S","e":"[telesales] [02:04 - 02:10] Dạ thì em biết là hồi trước giờ là như anh chia sẻ hồi nãy á thì em cũng hiểu là cũng lâu rồi anh cũng chưa mua sữa về cho ông bà nữa."},
{"id":2,"r":"S","e":"[telesales] [02:11 - 02:18] Đó vậy không biết là đợt này á là anh băn khoăn hay là mình lo ngại là sợ ông bà u... | [
{
"content": "You are a QC AI scoring the XỬ LÝ TỪ CHỐI skill of telesales staff.\nYour only job: read transcript → score 9 steps → output JSON only.\nDO NOT add any text outside JSON.\n\nCORE PRINCIPLE\nEvaluate TELESALE ACTION only — not customer response.\nTelesale acted correctly → S regardless of customer ... | Phân tích đoạn hội thoại telesales và trả về danh sách JSON các tình huống phản hồi theo từng lượt, gồm `id`, `r` (`S` hoặc `U`), `e` trích dẫn bằng tiếng Việt có mốc thời gian, và nếu là `U` thì thêm `g` là góp ý cải thiện ngắn gọn; tập trung vào việc nhận diện lời từ chối, giải pháp chốt đơn, và các câu hỏi mở còn th... |
C-117247501 | 70975920 | RTN | xulytuchoi | You are a QC AI scoring the XỬ LÝ TỪ CHỐI skill of telesales staff.
Your only job: read transcript → score 9 steps → output JSON only.
DO NOT add any text outside JSON.
CORE PRINCIPLE
Evaluate TELESALE ACTION only — not customer response.
Telesale acted correctly → S regardless of customer outcome.
ACTION > OUTCOME.
... | Module: xulytuchoi Transcript section: [customer] [02:39.000 - 02:40.500] Hôm bữa giảm 10% mà.
[telesales] [02:42.000 - 02:42.900] 10% hả?
[telesales] [02:44.000 - 02:49.200] Ừm, hồi đợt là mình mua ở trên đây luôn hả chị? Mua ở trên tổng đài của mấy bạn luôn đúng không?
[customer] [02:49.600 - 02:50.300] Đúng rồi em.
... | I need to analyze this transcript carefully for the XỬ LÝ TỪ CHỐI (objection handling) module.
Looking at the conversation flow:
- Customer mentions "Hôm bữa giảm 10% mà" - this seems like a comparison/mild objection about pricing
- Telesale checks the previous order details
- Telesale offers to keep the same price, f... | [
{"id":1,"r":"U","e":"[telesales] [02:42.000 - 02:42.900] '10% hả?' ; [telesales] [02:44.000 - 02:49.200] 'Ừm, hồi đợt là mình mua ở trên đây luôn hả chị?'","g":"Cải thiện: Cần thể hiện sự đồng cảm với khách hàng trước khi hỏi lại, ví dụ 'Dạ em hiểu, mình muốn được giá tốt như lần trước đúng không ạ' thay vì chỉ hỏi l... | [
{
"content": "You are a QC AI scoring the XỬ LÝ TỪ CHỐI skill of telesales staff.\nYour only job: read transcript → score 9 steps → output JSON only.\nDO NOT add any text outside JSON.\n\nCORE PRINCIPLE\nEvaluate TELESALE ACTION only — not customer response.\nTelesale acted correctly → S regardless of customer ... | Phân loại từng đoạn hội thoại telesales theo `r` (U/S) và giải thích ngắn gọn bằng tiếng Việt, trong đó nêu rõ đoạn nào cần cải thiện, đoạn nào là câu hỏi mở/làm rõ, đoạn nào đáp ứng yêu cầu giá + free ship + chốt đơn, và các đoạn còn lại ghi “Khách chốt đơn — Auto-S”. |
C-117667918 | 71636688 | RTN | xulytuchoi | You are a QC AI scoring the XỬ LÝ TỪ CHỐI skill of telesales staff.
Your only job: read transcript → score 9 steps → output JSON only.
DO NOT add any text outside JSON.
CORE PRINCIPLE
Evaluate TELESALE ACTION only — not customer response.
Telesale acted correctly → S regardless of customer outcome.
ACTION > OUTCOME.
... | Module: xulytuchoi Transcript section: [customer] [00:40] Nhưng mà cái đồ khuyến mãi của bên em á, cái máy xay cầm tay hôm trước chị lấy đồ khuyến mãi mà về cháu xay không được cái.
[telesales] [00:47] Ờ nó không xay được hả chị? Cái máy xay đó.
[customer] [00:50] Xoay vẫn xoay được. Xoay vẫn xoay được cái. Nạp điện vẫ... | I need to analyze this transcript carefully for the XỬ LÝ TỪ CHỐI (objection handling) skill.
**Context:** Customer's main objection is that a previous promotional gift (handheld blender) was defective, so she's lost trust in the brand's promotions and now buys from local stores instead. Source appears to be DGT/RTN b... | [
{"id":1,"r":"S","e":"[telesales] [00:47] 'Ờ nó không xay được hả chị? Cái máy xay đó.' ; [telesales] [01:02] 'chị quay lại giúp em một cái clip ngắn... để bộ phận chăm sóc khách hàng kiểm tra lại cái vấn đề này cho mình nha chị Thu' — Telesale thể hiện sự quan tâm và đề xuất giải pháp cho sự cố quà tặng"},
{"id":2,"r... | [
{
"content": "You are a QC AI scoring the XỬ LÝ TỪ CHỐI skill of telesales staff.\nYour only job: read transcript → score 9 steps → output JSON only.\nDO NOT add any text outside JSON.\n\nCORE PRINCIPLE\nEvaluate TELESALE ACTION only — not customer response.\nTelesale acted correctly → S regardless of customer ... | Phân loại từng lượt trao đổi trong đoạn hội thoại telesales–khách hàng thành `S` hoặc `U`, và với mỗi lượt hãy nêu ngắn gọn lý do cùng gợi ý cải thiện nếu là `U`. |
C-117729328 | 71960688 | RTN | xulytuchoi | You are a QC AI scoring the XỬ LÝ TỪ CHỐI skill of telesales staff.
Your only job: read transcript → score 9 steps → output JSON only.
DO NOT add any text outside JSON.
CORE PRINCIPLE
Evaluate TELESALE ACTION only — not customer response.
Telesale acted correctly → S regardless of customer outcome.
ACTION > OUTCOME.
... | Module: xulytuchoi Transcript section: [customer] [02:11.091 - 02:12.291] Dạ tại ở nhà còn á anh.
[telesales] [02:13.091 - 02:14.591] À dạ là nhà mình còn nhiều sữa vậy chị Nhi?
[customer] [02:15.591 - 02:19.291] Còn nhiều anh, tại vì em mới mua ở ngoài thêm một lon nữa.
[telesales] [02:19.991 - 02:22.991] À là một lon... | Looking at this transcript, I need to identify the rounds of objection handling.
**Round 1:**
- Customer objection: "còn nhiều anh, tại vì em mới mua ở ngoài thêm một lon nữa" (still has milk)
- Telesale responds with empathy, probing, and solution
**Round 2:**
- Customer objection: "mẹ em không chịu" / "mẹ ngán, khô... | [
{"id":1,"r":"S","e":"[Telesale] [02:59.191] 'Dạ nhưng mà em hiểu là nhà mình cũng còn sữa chị Nhi' — thể hiện sự đồng cảm và thấu hiểu tình trạng khách hàng còn sữa."},
{"id":2,"r":"S","e":"Auto-S — Khách nói 'còn nhiều anh, tại vì em mới mua ở ngoài thêm một lon nữa' thuộc exempt list (còn nhiều sữa / mới mua), khôn... | [
{
"content": "You are a QC AI scoring the XỬ LÝ TỪ CHỐI skill of telesales staff.\nYour only job: read transcript → score 9 steps → output JSON only.\nDO NOT add any text outside JSON.\n\nCORE PRINCIPLE\nEvaluate TELESALE ACTION only — not customer response.\nTelesale acted correctly → S regardless of customer ... | Read the sales-call transcript and classify each telesale utterance as S or U, with a brief Vietnamese explanation and quoted evidence for each label. |
C-117836926 | 72180169 | RTN | xulytuchoi | You are a QC AI scoring the XỬ LÝ TỪ CHỐI skill of telesales staff.
Your only job: read transcript → score 9 steps → output JSON only.
DO NOT add any text outside JSON.
CORE PRINCIPLE
Evaluate TELESALE ACTION only — not customer response.
Telesale acted correctly → S regardless of customer outcome.
ACTION > OUTCOME.
... | Module: xulytuchoi Transcript section: [customer] 01:32 [ 01:32.500 - 01:33.300] Dạ không ạ.
[telesales] 01:34 [ 01:34.400 - 01:34.700] Dạ.
[telesales] 01:35 [ 01:35.200 - 01:43.200] Chị thấy đợt này á là mình có lo ngại là mua nhiều quá cái uống không hết hay là có lo ngại là về chất lượng sữa hay sao nữa không chị? Đ... | Looking at this transcript, I need to identify the source data to determine if Urbox/CLP is required. No source info is explicitly given, so I'll evaluate based on what's present.
Let me map the 3 rounds of objection handling:
**Round 1:**
- Customer says "Dạ không ạ" (rejecting offer)
- Step 1 (Làm dịu): Telesale sa... | [
{"id":1,"r":"U","e":"[Telesale] [01:34] Dạ. — chỉ nói 'Dạ' mà không có câu thể hiện sự đồng cảm hay thấu hiểu trước từ chối của khách","g":"Cải thiện: Cần thể hiện sự đồng cảm trước khi hỏi thêm, ví dụ 'Dạ em hiểu ạ, chị chưa có nhu cầu thì cũng không sao ạ' trước khi chuyển sang làm rõ."},
{"id":2,"r":"S","e":"[Tele... | [
{
"content": "You are a QC AI scoring the XỬ LÝ TỪ CHỐI skill of telesales staff.\nYour only job: read transcript → score 9 steps → output JSON only.\nDO NOT add any text outside JSON.\n\nCORE PRINCIPLE\nEvaluate TELESALE ACTION only — not customer response.\nTelesale acted correctly → S regardless of customer ... | Phân tích ngắn gọn hội thoại telesale và trả về danh sách các câu/nói tương ứng với từng nhãn `U` hoặc `S`, kèm trích dẫn ngắn từ lời thoại và ghi chú lý do phân loại thật súc tích. |
C-117876003 | 71785374 | RTN | xulytuchoi | You are a QC AI scoring the XỬ LÝ TỪ CHỐI skill of telesales staff.
Your only job: read transcript → score 9 steps → output JSON only.
DO NOT add any text outside JSON.
CORE PRINCIPLE
Evaluate TELESALE ACTION only — not customer response.
Telesale acted correctly → S regardless of customer outcome.
ACTION > OUTCOME.
... | Module: xulytuchoi Transcript section: [customer] 00:49 - 00:49: Cái gì?
[telesales] 00:50 - 00:52: Dạ mua hai lon lớn sẽ tặng cho mình bộ hộp cơm giữ nhiệt đó.
[customer] 00:54 - 00:56: Hộp khuyến mại á hả?
[telesales] 00:57 - 00:59: À dạ đúng rồi, tặng kèm cho mình bộ hộp cơm á.
[customer] 01:01 - 01:02: Hộp cơm làm ... | Looking at this transcript, I need to identify the rounds of objection handling.
**Round 1:**
- Customer objection at 01:01-01:02: "Hộp cơm làm gì? Chả không cần." (doesn't need the gift)
- Telesale responds at 01:03-01:05 and 01:11-01:18 explaining the benefit of the gift
- Customer at 01:18-01:23: "có rồi, không khô... | [
{"id":1,"r":"U","e":"[Customer] 01:01 Hộp cơm làm gì? Chả không cần. ; [Telesale] 01:03 Dạ mình có thể đem cái tiện của mình đi.","g":"Cải thiện: Cần thể hiện sự đồng cảm trước khi phản biện, ví dụ: 'Dạ chú ơi, em hiểu ý chú, để em chia sẻ thêm cho chú nha'"},
{"id":2,"r":"U","e":"[Customer] 01:01 Hộp cơm làm gì? Chả... | [
{
"content": "You are a QC AI scoring the XỬ LÝ TỪ CHỐI skill of telesales staff.\nYour only job: read transcript → score 9 steps → output JSON only.\nDO NOT add any text outside JSON.\n\nCORE PRINCIPLE\nEvaluate TELESALE ACTION only — not customer response.\nTelesale acted correctly → S regardless of customer ... | Analyze the sales conversation and identify each turn’s coaching status as `S` or `U`, quoting the relevant dialogue and giving a short Vietnamese improvement suggestion for each `U` case. |
C-117888897 | 71823189 | RTN | xulytuchoi | You are a QC AI scoring the XỬ LÝ TỪ CHỐI skill of telesales staff.
Your only job: read transcript → score 9 steps → output JSON only.
DO NOT add any text outside JSON.
CORE PRINCIPLE
Evaluate TELESALE ACTION only — not customer response.
Telesale acted correctly → S regardless of customer outcome.
ACTION > OUTCOME.
... | Module: xulytuchoi Transcript section: [customer] 03:24 Dạ.
[telesales] 03:25 Dạ như lần này á chị.
[telesales] 03:28 Dạ?
[customer] 03:30 Em cũng suy nghĩ thêm.
[telesales] 03:32 Rồi dạ chị. Nghĩa là mình suy nghĩ lại vấn đề gì vậy chị?
[telesales] 03:36 Đợt này em thấy là cũng sắp lễ 30 tháng 4, 1 tháng 5 rồi á chị. ... | Looking at this transcript, I need to identify 3 rounds of objection handling (each with Làm dịu, Làm rõ, Làm hài lòng).
**Round 1:**
- Customer says "Em cũng suy nghĩ thêm" (objection)
- Telesale at 03:32: "Rồi dạ chị" → Làm dịu? Just acknowledgment, no empathy.
- Telesale at 03:32: "Nghĩa là mình suy nghĩ lại vấn đề... | [
{"id":1,"r":"U","e":"[telesales] 03:32 'Rồi dạ chị' — chỉ nói 'dạ chị' xác nhận, không thể hiện sự đồng cảm hay thấu hiểu cảm xúc khách hàng khi khách nói 'suy nghĩ thêm'.","g":"Cải thiện: Cần thể hiện sự đồng cảm thực sự, ví dụ 'Dạ em hiểu chị muốn suy nghĩ thêm ạ, điều đó hoàn toàn hợp lý ạ'."},
{"id":2,"r":"S","e"... | [
{
"content": "You are a QC AI scoring the XỬ LÝ TỪ CHỐI skill of telesales staff.\nYour only job: read transcript → score 9 steps → output JSON only.\nDO NOT add any text outside JSON.\n\nCORE PRINCIPLE\nEvaluate TELESALE ACTION only — not customer response.\nTelesale acted correctly → S regardless of customer ... | Identify which telesales utterances are effective vs ineffective objection-handling, and briefly explain why each one is labeled that way. |
C-117954910 | 72159268 | RTN | xulytuchoi | "You are a QC AI scoring the XỬ LÝ TỪ CHỐI skill of telesales staff.\nYour only job: read tra(...TRUNCATED) | "Module: xulytuchoi Transcript section: [customer] 02:37: Khi nào chị có nhu cầu thì chị li(...TRUNCATED) | "Looking at this transcript, I need to identify the customer's objections and map the telesale's res(...TRUNCATED) | "[\n{\"id\":1,\"r\":\"S\",\"e\":\"[Telesale] [02:40] em cũng hiểu là giờ cô còn sữa cho n(...TRUNCATED) | [{"content":"You are a QC AI scoring the XỬ LÝ TỪ CHỐI skill of telesales staff.\nYour only j(...TRUNCATED) | "Phân tích đoạn hội thoại telesale theo từng mốc thời gian, gán nhãn S/U cho từng(...TRUNCATED) |
C-118080909 | 72252355 | RTN | xulytuchoi | "You are a QC AI scoring the XỬ LÝ TỪ CHỐI skill of telesales staff.\nYour only job: read tra(...TRUNCATED) | "Module: xulytuchoi Transcript section: [customer] [02:40 - 02:49] Tại vì lâu lâu mà mẹ bả(...TRUNCATED) | "Looking at this transcript, I need to identify the customer's objections and map the telesale's res(...TRUNCATED) | "[\n{\"id\":1,\"r\":\"S\",\"e\":\"[Telesale] [02:50-02:53] 'Dạ, nhưng mà mẹ dùng Ensure đư(...TRUNCATED) | [{"content":"You are a QC AI scoring the XỬ LÝ TỪ CHỐI skill of telesales staff.\nYour only j(...TRUNCATED) | "Đánh giá transcript telesale theo từng đoạn, gán nhãn `S` hoặc `U` kèm giải thích n(...TRUNCATED) |
C-118081368 | 72365296 | RTN | xulytuchoi | "You are a QC AI scoring the XỬ LÝ TỪ CHỐI skill of telesales staff.\nYour only job: read tra(...TRUNCATED) | "Module: xulytuchoi Transcript section: [customer] 01:42 À có gì thì chị chị sẽ liên lạ(...TRUNCATED) | "Looking at this transcript, I need to identify 3 rounds of objection handling.\n\n**Round 1:** Cust(...TRUNCATED) | "[\n{\"id\":1,\"r\":\"S\",\"e\":\"[Telesale] [01:47] 'chị nghe em cũng hiểu là chị cũng mu(...TRUNCATED) | [{"content":"You are a QC AI scoring the XỬ LÝ TỪ CHỐI skill of telesales staff.\nYour only j(...TRUNCATED) | "Analyze the telesale transcript and return a numbered list of positive/negative evidence entries wi(...TRUNCATED) |
End of preview. Expand in Data Studio
200-Sample-T2
Made with ❤️ using 🦥 Unsloth Studioxin xin was generated with Unsloth Recipe Studio. It contains 200 generated records.
🚀 Quick Start
from datasets import load_dataset
# Load the main dataset
dataset = load_dataset("Nhan1006/200-sample-t2", "data", split="train")
df = dataset.to_pandas()
📊 Dataset Summary
- 📈 Records: 200
- 📋 Columns: 10
📋 Schema & Statistics
| Column | Type | Column Type | Unique (%) | Null (%) | Details |
|---|---|---|---|---|---|
generated_instruction |
string |
llm-text | 200 (100.0%) | 0 (0.0%) | Tokens: 86 out / 1079 in |
⚙️ Generation Details
Generated with 10 column configuration(s):
llm-text: 1 column(s)
seed-dataset: 9 column(s)
📄 Full configuration available in builder_config.json and detailed metadata in metadata.json.
📚 Citation
If you use Data Designer in your work, please cite the project as follows:
@misc{nemo-data-designer,
author = {The NeMo Data Designer Team, NVIDIA},
title = {NeMo Data Designer: A framework for generating synthetic data from scratch or based on your own seed data},
howpublished = {\url{https://github.com/NVIDIA-NeMo/DataDesigner}},
year = 2026,
note = {GitHub Repository},
}
💡 About NeMo Data Designer
NeMo Data Designer is a general framework for generating high-quality synthetic data that goes beyond simple LLM prompting. It provides:
- Diverse data generation using statistical samplers, LLMs, or existing seed datasets
- Relationship control between fields with dependency-aware generation
- Quality validation with built-in Python, SQL, and custom local and remote validators
- LLM-as-a-judge scoring for quality assessment
- Fast iteration with preview mode before full-scale generation
For more information, visit: https://github.com/NVIDIA-NeMo/DataDesigner (pip install data-designer)
- Downloads last month
- 42