Author

Zuoyebang Tech Team

Sharing technical practices from Zuoyebang

Articles

Likes

Views

Comments

Latest from Zuoyebang Tech Team

40 recent articles

Zuoyebang Tech Team

May 26, 2022 · Cloud Native

How ZRTC Powers Millions of Live Streams: Architecture & Scaling

ZRTC, the real‑time audio‑video platform behind 作业帮, has been refined for over three years to support massive, multi‑cloud, multi‑protocol live streaming, employing a unified SDK, intelligent scheduling, custom SFU services, and extensive performance tuning to achieve high concurrency, low latency, and robust high‑availability.

High ConcurrencyReal-time StreamingSystem Architecture

0 likes · 22 min read

How ZRTC Powers Millions of Live Streams: Architecture & Scaling

Zuoyebang Tech Team

May 19, 2022 · Artificial Intelligence

How to Achieve High‑Quality TTS with Only Minutes of Data

This article reviews neural speech synthesis, explains why large high‑quality paired data are essential, and presents a range of low‑resource solutions—including semi‑supervised pre‑training, cross‑language transfer, speaker embedding, and Conformer‑based model upgrades—demonstrating how the Zuoyebang team built a robust TTS system with as little as 7‑minute speaker recordings.

ConformerFastspeech2low-resource TTS

0 likes · 15 min read

How to Achieve High‑Quality TTS with Only Minutes of Data

Zuoyebang Tech Team

May 13, 2022 · Operations

Build a Scalable, Cost‑Effective Log Retrieval System Without Elasticsearch

This article explains how to design a high‑performance, low‑cost log retrieval architecture that avoids Elasticsearch by partitioning logs into time‑based chunks, indexing only metadata, using multi‑tier storage (local, remote, archive), and orchestrating queries through GD‑Search, Local‑Search, Remote‑Search and Log‑Manager components.

cost efficiencydistributed systemslog retrieval

0 likes · 14 min read

Build a Scalable, Cost‑Effective Log Retrieval System Without Elasticsearch

Zuoyebang Tech Team

May 9, 2022 · Big Data

How Flink SQL Powered Real‑Time Learning Analytics at Zuoyebang

Zuoyebang’s big‑data team shares how they evolved from SparkStreaming to a Flink‑SQL‑centric real‑time platform, detailing three development stages, challenges in DAG optimization, Redis‑based table design, and platform features for unified deployment, ease of use, and operational governance.

FlinkSQLStreaming

0 likes · 14 min read

How Flink SQL Powered Real‑Time Learning Analytics at Zuoyebang

Zuoyebang Tech Team

Apr 29, 2022 · Artificial Intelligence

How AI-Powered Speech Evaluation Transforms Language Learning

This article explains the background, evaluation metrics, and technical framework of computer‑assisted language learning, detailing how modern AI‑driven speech assessment systems use HMM‑DNN models, GOP scoring, conformer architectures, streaming solutions, and edge‑cloud deployment to deliver accurate, low‑latency pronunciation feedback for learners.

AIDNNEdge computing

0 likes · 14 min read

How AI-Powered Speech Evaluation Transforms Language Learning

Zuoyebang Tech Team

Apr 26, 2022 · Cloud Native

How Serverless Kubernetes Virtual Nodes Cut Costs and Boost Scalability

Zhang's team at Zuoyebang details their journey to serverless Kubernetes virtual nodes, explaining how elastic scaling, fine-grained scheduling, and cost‑effective resource utilization transformed high‑peak online services, while addressing challenges in scheduling, observability, performance, and multi‑cloud resilience.

Kubernetescost optimizationresource scheduling

0 likes · 11 min read

How Serverless Kubernetes Virtual Nodes Cut Costs and Boost Scalability

Zuoyebang Tech Team

Apr 21, 2022 · Backend Development

How to Build a High‑Concurrency, Low‑Latency Live Streaming System for Online Education

This article details the design and implementation of a self‑developed interactive live‑streaming platform that supports massive concurrent users and ultra‑low latency for online education, covering business scenarios, technical abstractions, key low‑latency and high‑concurrency techniques, and real‑world performance results.

High ConcurrencyKCPWebRTC

0 likes · 16 min read

How to Build a High‑Concurrency, Low‑Latency Live Streaming System for Online Education

Zuoyebang Tech Team

Apr 15, 2022 · Artificial Intelligence

Zuoyebang’s NLP Platforms: Boosting Online Education with AI

In this interview, Zuoyebang’s NLP lead explains how the company built self‑developed platforms like IQC and FTP to automate text quality inspection and intelligent labeling, outlines their architecture, shares practical deep‑learning applications such as translation and grammar correction, and discusses future research directions in large‑scale multi‑label classification, few‑shot learning, and multimodal models.

AI platformsNLPmachine learning

0 likes · 11 min read

Zuoyebang’s NLP Platforms: Boosting Online Education with AI

Zuoyebang Tech Team

Apr 13, 2022 · Big Data

How Delta Lake Transformed Our Offline Data Warehouse Performance

This article details how ZuoYeBang's engineering team migrated their Hive‑based offline data warehouse to Delta Lake, tackling latency, scalability, and query‑performance challenges through stream‑to‑batch processing, data‑lake architecture, and optimizations like DPP and Z‑ordering.

Big DataDelta LakeHive

0 likes · 15 min read

How Delta Lake Transformed Our Offline Data Warehouse Performance

Zuoyebang Tech Team

Apr 7, 2022 · Cloud Native

How Fluid Transforms Large‑Scale Data Retrieval on Kubernetes

This article explains how Zuoyebang redesigned its massive data retrieval platform by separating compute and storage with the Fluid project on Kubernetes, achieving minute‑level hundred‑TB distribution, elastic caching, and improved stability for real‑time educational services.

Compute-Storage SeparationData RetrievalFluid

0 likes · 8 min read

How Fluid Transforms Large‑Scale Data Retrieval on Kubernetes