---
language:
- zh
tags:
- hisilicon
- hispark
- npu
- openharmony
- modelzoo
- pytorch
---

# FastSpeech2

FastSpeech2 是一种高效的端到端语音合成模型。相比 FastSpeech，FastSpeech2 引入了多尺度时长预测器和能量 / 基频预测分支，优化了时长预测模块并新增韵律特征建模，在合成速度和语音自然度上均有大幅提升。

## Mirror Metadata

- Hugging Face repo: shadow-cann/hispark-modelzoo-fastspeech2
- Portal model id: ie2sc9g1qk00
- Created at: 2026-01-08 16:18:57
- Updated at: 2026-01-09 15:08:50
- Category: 音频

## Framework

- PyTorch

## Supported OS

- OpenHarmony
- Linux

## Computing Power

- Hi3403V100 SVP_NNN

## Tags

- 文本转语音

## Detail Parameters

- 输入: 1x40
- 参数量: 35.266M
- 计算量: 29.162GFLOPs

## Files In This Repo

- fastspeech_hifigan_en.onnx (源模型 / 源模型下载; 源模型 / 源模型元数据; 编译模型 / OM 元数据 / a16w8)

## Upstream Links

- Portal card: https://gitbubble.github.io/hisilicon-developer-portal-mirror/model-detail.html?id=ie2sc9g1qk00
- Upstream repository: https://gitee.com/HiSpark/modelzoo/blob/master/samples/built-in/audio/FastSpeech2/README.md
- License reference: https://github.com/ming024/FastSpeech2/blob/master/LICENSE

## Notes

- This repository was mirrored from the HiSilicon Developer Portal model card and local downloads captured on 2026-03-27.
- File ownership follows the portal card mapping, not just filename similarity.
- Cover image: 1722265270026243_fastspeech2.jpg