update README
This commit is contained in:
1 parent
5088f0b50a
commit
ff78032400
603 files changed
+21
-23
No files matched your search
@@ -6,7 +6,7 @@ GPT-SoVITS 模型是一个集成了语音转换和文本转语音功能的先进
|
||||
Kokoro 是由 hexgrad 团队开发并开源的轻量级、高性能文本转语音(TTS)模型。
|
||||
F5-TTS 模型由上海交通大学团队发布,是基于扩散 Transformer 和 ConvNeXt V2 的文本转语音(TTS)模型。
|
||||
|
||||
## GPT-SoVITS 模型测试服务原理
|
||||
<!-- ## GPT-SoVITS 模型测试服务原理
|
||||
|
||||
使用 GPT-SoVITS 框架内置的类 TTS 和 TTS_Config。
|
||||
其中 TTS_Config 封装了 GPT-SoVITS 模型运行所需的全部配置参数;TTS 类是 GPT-SoVITS 模型的高层封装,整合了模型加载、文本处理、语音生成等全流程逻辑。
|
||||
@@ -107,16 +107,18 @@ if response.status_code == 200:
|
||||
print("音频保存成功:output.pcm")
|
||||
else:
|
||||
print(f"请求失败:{response.status_code},{response.text}")
|
||||
```
|
||||
``` -->
|
||||
|
||||
## 如何使用语音合成模型测试框架
|
||||
代码实现了一个接收音频数据并返回识别文本的语音识别 HTTP 服务,并基于 corex:3.2.1 基础镜像,将该 HTTP 服务重新打包成 docker 镜像,通过 k8s 集群sut容器去请求这个 HTTP 服务。
|
||||
代码实现了一个接收音频数据并返回识别文本的语音识别 HTTP 服务,将该 HTTP 服务重新打包成 docker 镜像,通过 k8s 集群sut容器去请求这个 HTTP 服务。
|
||||
|
||||
## 寒武纪MLU370系列上语音合成模型运行测试结果
|
||||
在寒武纪MLU370系列上对部分语音合成模型进行适配,测试方式为在 Nvidia A100 和 寒武纪MLU370系列加速卡上对同一段text进行语音合成任务,获取运行时间
|
||||
|
||||
| 模型名称 | 模型类型 | 适配状态 | 天垓100运行时间/s | Nvidia A100运行时间/s |
|
||||
| ---------- | ---------------------- | -------- | ----------------- | --------------------- |
|
||||
| kokoro | StyleTTS 2, ISTFTNet | 成功 | 0.2 | 0.1 |
|
||||
| f5-TTS | DiT, ConvNeXt V2 | 成功 | 0.5 | 0.2 |
|
||||
| gpt-sovits | VITS | 成功 | 0.9 | 0.4 |
|
||||
| 模型名称 | 模型类型 | 适配状态 | 寒武纪MLU370-X8运行时间/s | 寒武纪MLU370-X4运行时间/s | Nvidia A100运行时间/s |
|
||||
| ---------- | ---------------------- | -------- | ----------------- | --------------------- | --------------------- |
|
||||
| kokoro | StyleTTS 2, ISTFTNet | 成功 | 2.5 | 2.2 | 5.4 |
|
||||
| f5-TTS | DiT, ConvNeXt V2 | 成功 | 39.1 | 32.1 | 5.4 |
|
||||
| gpt-sovits | VITS | 成功 | 28.8 | 27.2 | 20.5 |
|
||||
| matcha | OT-CFM, Transformer | 成功 | 2.5 | 2.1 | 3.1 |
|
||||
| piper | - | 成功 | 0.8 | 0.7 | 1.9 |
|
||||
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
@@ -15,8 +15,8 @@ docker build -t tts:f5 . -f Dockerfile_f5
|
||||
```bash
|
||||
docker run -it --rm \
|
||||
-v /models/F5-TTS_Emilia-ZH-EN:/mnt/models \
|
||||
-v /dev:/dev \
|
||||
--device=/dev/iluvatar0:/dev/iluvatar0 \
|
||||
--device=/dev/cambricon_dev0:/dev/cambricon_dev0 \
|
||||
--device=/dev/cambricon_ctl:/dev/cambricon_ctl \
|
||||
-p 8080:80 \
|
||||
-e MODEL_DIR=/mnt/models \
|
||||
-e MODEL_NAME=model_1250000.safetensors \
|
||||
@@ -27,7 +27,7 @@ docker run -it --rm \
|
||||
- `MODEL_DIR`:模型所在目录(挂载到容器内 `/mnt/models`)
|
||||
- `MODEL_NAME`:加载的模型文件名(通常为 `.safetensors`)
|
||||
- `-p 8080:80`:将容器内服务端口映射到宿主机 `8080`
|
||||
- `--device=/dev/iluvatar0:/dev/iluvatar0`:指定推理设备(如 GPU/加速卡)
|
||||
|
||||
|
||||
### 3. 测试服务
|
||||
```bash
|
||||
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
File renamed without changes.
Loaded 100 of 603 files, more files were not shown because too many files have changed in this diff.
Show more
Reference in new issue
Block a user