AssemblyAI 用 Amazon EC2 Spot 优化语音转文字推理，成本降低 70%
Cost-optimizing AssemblyAI speech-to-text inference on AWS with Amazon EC2 Spot
AWS 官方拆解了 AssemblyAI 的推理降本方案：用 Spot GPU 实例加 SQS 重试队列，省下 70% 成本还不破坏 SLA。跑语音转写或 GPU 推理服务的团队可以直接照搬这套容错架构，比单纯换机型省得多。
AssemblyAI 在 AWS 上用 Amazon EC2 Spot GPU 实例运行语音转文字（speech-to-text）推理，实现约 70% 的成本节省，同时维持面向欧洲用户的 SLA 承诺。该方案的核心组件包括 Amazon ECS、Spot GPU 实例和基于 Amazon SQS 的重试架构。
技术上，任务通过 Amazon SQS 队列分发，Spot 实例被回收时，未完成的任务会回到队列由其他实例重新处理，配合 ECS 的弹性伸缩应对 Spot 中断，从而在不牺牲可用性的前提下用低价 Spot 容量替代按需 GPU 实例。
这套架构覆盖多个欧洲 Region，对需要大规模跑语音识别推理的团队有直接参考价值：用队列加重试的方式消化 Spot 中断，是 GPU 推理降本可复用的模式。
Cost-optimizing AssemblyAI speech-to-text inference on AWS with Amazon EC2 Spot
Learn how AssemblyAI runs cost-optimized speech-to-text inference at scale on AWS, using Amazon ECS, Amazon EC2 Spot GPU instances, and an Amazon SQS-based retry architecture to achieve 70% cost savings while maintaining SLA commitments across European Regions.
This report is categorized under AITOP's "How-to" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include AssemblyAI, Amazon EC2 Spot, AWS, speech-to-text.
Primary source verification: published directly by AWS Machine Learning Blog with first-party authority.
