ModelScope 发布 v1.40.1：新增 JAEC 推理与语音分离模型

AI 摘要

阿里 ModelScope 发布 v1.40.1 版本，新增 JAEC 原生推理流水线以及 FLA-TF-Locoformer、FLA-T-SepReformer 两个语音分离模型，并加入 Agent IDP 与 MCP/Studio OpenAPI 支持。同时修复了 vLLM Ascend 在 OpenEuler 和 x86 上的构建依赖问题，统一上传环境变量，并升级 modelscope-hub 以统一 CLI 与认证逻辑。

正文节选

## Release Notes ### Features - **Add JAEC native inference pipeline** — Introduced a native inference pipeline for Joint Acoustic Echo Cancellation (JAEC) under the audio/AEC module. - **Add FLA-TF-Locoformer and FLA-T-SepReformer separators** — Added two new speech separation models (FLA-TF-Locoformer and FLA-T-SepReformer) for 2-speaker 8 kHz audio, including shared attention layers, pipeline registration, test cases, and ONNX export support. - **Add Agent IDP and MCP/Studio OpenAPI**

发布时间：2026-09-15 23:47

抓取时间：2026-09-15 23:58

来源机构：魔搭社区