github/datascale-ai-opentalking
1
0
Fork 0
mirror of https://github.com/datascale-ai/opentalking.git synced 2026-07-05 08:31:24 +08:00
Code Issues Packages Projects Releases Wiki Activity

docs: refresh README documentation links (#129)

Browse Source
This commit is contained in:
zyairehhh
2026-06-29 19:54:17 +08:00
committed by GitHub
parent 5473bb2665
commit d3083e799d
3 changed files with 66 additions and 66 deletions
Download Patch File Download Diff File Expand all files Collapse all files

44
README.en.md
View File

@@ -123,12 +123,12 @@ OpenTalking's **orchestration layer** (API / Worker / frontend) and **digital-hu
| Path | Recommended model / backend | Device reference | Best for | Details |
| --- | --- | --- | --- | --- |
| Fast trial | `mock` | CPU / no GPU | Validate API, LLM, TTS, WebRTC, and browser playback without downloading model weights | [Quickstart](docs/en/user-guide/quickstart.md) |
| Entry validation | `quicktalk` / `wav2lip` | RTX 3050 Laptop, RTX 3060, RTX 4060 | Run real video rendering for demos and deployment validation; lower the resolution on low-memory devices | [QuickTalk](docs/en/model-deployment/quicktalk.md) / [Wav2Lip](docs/en/model-deployment/wav2lip-local.md) |
| Consumer-GPU single machine | `quicktalk` / `wav2lip` / `musetalk` | RTX 3090, RTX 4090 | Closer to real-time local demos, private validation, and lightweight pre-production evaluation | [Model deployment](docs/en/model-deployment/index.md) |
| Fully local private path | `sensevoice` + `local_cosyvoice` + `quicktalk` | RTX 3090 / 4090 or similar GPU | Run STT, TTS, and video driving locally; OpenTalking uses the main `.venv`, while CosyVoice runs in a dedicated sidecar venv | [Local STT/TTS + QuickTalk](docs/en/model-deployment/local-quicktalk-audio.md) |
| High-quality remote inference | `flashtalk` / `flashhead` / `fasterliveportrait` + OmniRT | Multi-GPU, Ascend 910B2, remote GPU service | Multi-card, GPU/NPU, production isolation, higher visual quality, or video clone workflows | [FlashTalk](docs/en/model-deployment/flashtalk.md) / [FasterLivePortrait](docs/en/model-deployment/fasterliveportrait.md) |
| Docker / production deployment | API, Web, Worker, external model services | Single GPU, remote GPU, distributed cluster | Service deployment, remote GPU, distributed runtime, and production validation | [Deployment](docs/en/user-guide/deployment.md) |
| Fast trial | `mock` | CPU / no GPU | Validate API, LLM, TTS, WebRTC, and browser playback without downloading model weights | [Quickstart](https://datascale-ai.github.io/opentalking/latest/en/quick-start/) |
| Entry validation | `quicktalk` / `wav2lip` | RTX 3050 Laptop, RTX 3060, RTX 4060 | Run real video rendering for demos and deployment validation; lower the resolution on low-memory devices | [QuickTalk](https://datascale-ai.github.io/opentalking/latest/en/avatar_models/deployment/quicktalk-local/) / [Wav2Lip](https://datascale-ai.github.io/opentalking/latest/en/avatar_models/deployment/wav2lip-local/) |
| Consumer-GPU single machine | `quicktalk` / `wav2lip` / `musetalk` | RTX 3090, RTX 4090 | Closer to real-time local demos, private validation, and lightweight pre-production evaluation | [Model and backend selection](https://datascale-ai.github.io/opentalking/latest/en/model-support/selection/) |
| Fully local private path | `sensevoice` + `local_cosyvoice` + `quicktalk` | RTX 3090 / 4090 or similar GPU | Run STT, TTS, and video driving locally; OpenTalking uses the main `.venv`, while CosyVoice runs in a dedicated sidecar venv | [Local STT/TTS + QuickTalk](https://datascale-ai.github.io/opentalking/latest/en/recipes/local-quicktalk-audio/) |
| High-quality remote inference | `flashtalk` / `flashhead` / `fasterliveportrait` + OmniRT | Multi-GPU, Ascend 910B2, remote GPU service | Multi-card, GPU/NPU, production isolation, higher visual quality, or video clone workflows | [FlashTalk](https://datascale-ai.github.io/opentalking/latest/en/avatar_models/flashtalk/) / [FasterLivePortrait](https://datascale-ai.github.io/opentalking/latest/en/avatar_models/fasterliveportrait/) |
| Docker / production deployment | API, Web, Worker, external model services | Single GPU, remote GPU, distributed cluster | Service deployment, remote GPU, distributed runtime, and production validation | [Deployment](https://datascale-ai.github.io/opentalking/latest/en/deployment/) |
## Quickstart
@@ -145,7 +145,7 @@ If you want to try the OpenTalking + OmniRT + QuickTalk real-time digital-human
- Image URL: [image link](https://www.compshare.cn/images/TdDwmKZUZebI?referral_code=Hid5KUhcqlZEptmMEwKy2F)
- Exposed port: `5173` (WebUI; API traffic is proxied internally)
- Guide: [Compshare image quick experience](docs/en/quick-start/compshare-image.md)
- Guide: [Compshare image quick experience](https://datascale-ai.github.io/opentalking/latest/en/quick-start/)
The image includes OpenTalking, OmniRT, the QuickTalk runtime environment, and model files. After deploying an instance, open port `5173` and visit the instance URL provided by the platform. If you need to restart services manually, follow the commands in the guide.
@@ -162,7 +162,7 @@ source .venv/bin/activate
cp .env.example .env
```
Edit `.env` and configure at least an LLM. The default TTS can use the keyless `edge` voice. LLM, STT, and TTS are independent providers; see [Configuration](docs/en/user-guide/configuration.md) and [LLM / STT](docs/en/model-deployment/llm-stt.md).
Edit `.env` and configure at least an LLM. The default TTS can use the keyless `edge` voice. LLM, STT, and TTS are independent providers; see [Configuration](https://datascale-ai.github.io/opentalking/latest/en/reference/configuration/) and [LLM / STT](https://datascale-ai.github.io/opentalking/latest/en/speech_models/llm-stt/).
```bash
bash scripts/start_unified.sh --mock
@@ -202,12 +202,12 @@ bash scripts/start_unified.sh \
More entrypoints:
- [QuickTalk local deployment](docs/en/model-deployment/quicktalk.md)
- [Wav2Lip local deployment](docs/en/model-deployment/wav2lip-local.md)
- [FasterLivePortrait / JoyVASA](docs/en/model-deployment/fasterliveportrait.md)
- [Video clone guide](docs/en/usage/webui/video-clone.md)
- [WebUI guide](docs/en/usage/webui/basic.md)
- [Docker Compose and production deployment](docs/en/user-guide/deployment.md)
- [QuickTalk local deployment](https://datascale-ai.github.io/opentalking/latest/en/avatar_models/deployment/quicktalk-local/)
- [Wav2Lip local deployment](https://datascale-ai.github.io/opentalking/latest/en/avatar_models/deployment/wav2lip-local/)
- [FasterLivePortrait / JoyVASA](https://datascale-ai.github.io/opentalking/latest/en/avatar_models/fasterliveportrait/)
- [Video clone guide](https://datascale-ai.github.io/opentalking/latest/en/usage/webui/video-clone/)
- [WebUI guide](https://datascale-ai.github.io/opentalking/latest/en/usage/webui/basic/)
- [Docker Compose and production deployment](https://datascale-ai.github.io/opentalking/latest/en/deployment/)
## Supported Models
@@ -227,7 +227,7 @@ More entrypoints:
| --- | --- | --- | --- | --- | --- |
| `quicktalk` | RTX 3090 | Template video + audio | 720x900 / 25fps | About 3.8 GiB | About 35 fps |
For weight downloads, Docker, troubleshooting, and model configuration, see [Model deployment](docs/en/model-deployment/index.md).
For weight downloads, Docker, troubleshooting, and model configuration, see [Model deployment](https://datascale-ai.github.io/opentalking/latest/en/model-deployment/).
### Cloud Model API: Atlas Cloud
@@ -239,7 +239,7 @@ For weight downloads, Docker, troubleshooting, and model configuration, see [Mod
> **[Atlas Cloud](https://www.atlascloud.ai/?utm_source=github&utm_medium=link&utm_campaign=opentalking)** is an all-modal AI inference platform. One API gives you access to video generation, image generation, and LLMs, so you do not need to integrate multiple vendors separately. A single integration can route to 300+ curated all-modal models.
OpenTalking uses an OpenAI-compatible interface for LLMs. Point `OPENTALKING_LLM_BASE_URL` to `https://api.atlascloud.ai/v1` to use Atlas-hosted DeepSeek / Qwen models. See [LLM and STT](docs/en/model-deployment/llm-stt.md). For budget-friendly API options, see Atlas Cloud's [coding plan](https://www.atlascloud.ai/console/coding-plan).
OpenTalking uses an OpenAI-compatible interface for LLMs. Point `OPENTALKING_LLM_BASE_URL` to `https://api.atlascloud.ai/v1` to use Atlas-hosted DeepSeek / Qwen models. See [LLM and STT](https://datascale-ai.github.io/opentalking/latest/en/speech_models/llm-stt/). For budget-friendly API options, see Atlas Cloud's [coding plan](https://www.atlascloud.ai/console/coding-plan).
## Progress And Roadmap
@@ -304,12 +304,12 @@ OpenTalking uses an OpenAI-compatible interface for LLMs. Point `OPENTALKING_LLM
## Documentation And Community
- [Quickstart](docs/en/user-guide/quickstart.md)
- [Models](docs/en/model-deployment/index.md) (weight downloads, mirrors, startup, validation)
- [Architecture](docs/en/developer-guide/architecture.md)
- [Configuration](docs/en/user-guide/configuration.md)
- [Deployment](docs/en/user-guide/deployment.md) (Docker Compose, distributed deployment)
- [Model adapter](docs/en/developer-guide/model-adapter.md)
- [Quickstart](https://datascale-ai.github.io/opentalking/latest/en/quick-start/)
- [Models](https://datascale-ai.github.io/opentalking/latest/en/model-deployment/) (weight downloads, mirrors, startup, validation)
- [Architecture](https://datascale-ai.github.io/opentalking/latest/en/developer-guide/architecture/)
- [Configuration](https://datascale-ai.github.io/opentalking/latest/en/reference/configuration/)
- [Deployment](https://datascale-ai.github.io/opentalking/latest/en/deployment/) (Docker Compose, distributed deployment)
- [Model adapter](https://datascale-ai.github.io/opentalking/latest/en/developer-guide/model-adapter/)
- [Contributing](CONTRIBUTING.md) (dev environment, CLI tools, ruff / mypy / pytest)
Join the QQ community to discuss real-time digital humans, FlashTalk, OmniRT, model deployment, and product scenarios.

44
README.md
View File

@@ -123,12 +123,12 @@ OpenTalking's **orchestration layer** (API / Worker / frontend) and **digital-hu
| Path | Recommended model / backend | Device reference | Best for | Details |
| --- | --- | --- | --- | --- |
| Fast trial | `mock` | CPU / no GPU | Validate API, LLM, TTS, WebRTC, and browser playback without downloading model weights | [Quickstart](docs/en/user-guide/quickstart.md) |
| Entry validation | `quicktalk` / `wav2lip` | RTX 3050 Laptop, RTX 3060, RTX 4060 | Run real video rendering for demos and deployment validation; lower the resolution on low-memory devices | [QuickTalk](docs/en/model-deployment/quicktalk.md) / [Wav2Lip](docs/en/model-deployment/wav2lip-local.md) |
| Consumer-GPU single machine | `quicktalk` / `wav2lip` / `musetalk` | RTX 3090, RTX 4090 | Closer to real-time local demos, private validation, and lightweight pre-production evaluation | [Model deployment](docs/en/model-deployment/index.md) |
| Fully local private path | `sensevoice` + `local_cosyvoice` + `quicktalk` | RTX 3090 / 4090 or similar GPU | Run STT, TTS, and video driving locally; OpenTalking uses the main `.venv`, while CosyVoice runs in a dedicated sidecar venv | [Local STT/TTS + QuickTalk](docs/en/model-deployment/local-quicktalk-audio.md) |
| High-quality remote inference | `flashtalk` / `flashhead` / `fasterliveportrait` + OmniRT | Multi-GPU, Ascend 910B2, remote GPU service | Multi-card, GPU/NPU, production isolation, higher visual quality, or video clone workflows | [FlashTalk](docs/en/model-deployment/flashtalk.md) / [FasterLivePortrait](docs/en/model-deployment/fasterliveportrait.md) |
| Docker / production deployment | API, Web, Worker, external model services | Single GPU, remote GPU, distributed cluster | Service deployment, remote GPU, distributed runtime, and production validation | [Deployment](docs/en/user-guide/deployment.md) |
| Fast trial | `mock` | CPU / no GPU | Validate API, LLM, TTS, WebRTC, and browser playback without downloading model weights | [Quickstart](https://datascale-ai.github.io/opentalking/latest/en/quick-start/) |
| Entry validation | `quicktalk` / `wav2lip` | RTX 3050 Laptop, RTX 3060, RTX 4060 | Run real video rendering for demos and deployment validation; lower the resolution on low-memory devices | [QuickTalk](https://datascale-ai.github.io/opentalking/latest/en/avatar_models/deployment/quicktalk-local/) / [Wav2Lip](https://datascale-ai.github.io/opentalking/latest/en/avatar_models/deployment/wav2lip-local/) |
| Consumer-GPU single machine | `quicktalk` / `wav2lip` / `musetalk` | RTX 3090, RTX 4090 | Closer to real-time local demos, private validation, and lightweight pre-production evaluation | [Model and backend selection](https://datascale-ai.github.io/opentalking/latest/en/model-support/selection/) |
| Fully local private path | `sensevoice` + `local_cosyvoice` + `quicktalk` | RTX 3090 / 4090 or similar GPU | Run STT, TTS, and video driving locally; OpenTalking uses the main `.venv`, while CosyVoice runs in a dedicated sidecar venv | [Local STT/TTS + QuickTalk](https://datascale-ai.github.io/opentalking/latest/en/recipes/local-quicktalk-audio/) |
| High-quality remote inference | `flashtalk` / `flashhead` / `fasterliveportrait` + OmniRT | Multi-GPU, Ascend 910B2, remote GPU service | Multi-card, GPU/NPU, production isolation, higher visual quality, or video clone workflows | [FlashTalk](https://datascale-ai.github.io/opentalking/latest/en/avatar_models/flashtalk/) / [FasterLivePortrait](https://datascale-ai.github.io/opentalking/latest/en/avatar_models/fasterliveportrait/) |
| Docker / production deployment | API, Web, Worker, external model services | Single GPU, remote GPU, distributed cluster | Service deployment, remote GPU, distributed runtime, and production validation | [Deployment](https://datascale-ai.github.io/opentalking/latest/en/deployment/) |
## Quickstart
@@ -145,7 +145,7 @@ If you want to try the OpenTalking + OmniRT + QuickTalk real-time digital-human
- Image URL: [image link](https://www.compshare.cn/images/TdDwmKZUZebI?referral_code=Hid5KUhcqlZEptmMEwKy2F)
- Exposed port: `5173` (WebUI; API traffic is proxied internally)
- Guide: [Compshare image quick experience](docs/en/quick-start/compshare-image.md)
- Guide: [Compshare image quick experience](https://datascale-ai.github.io/opentalking/latest/en/quick-start/)
The image includes OpenTalking, OmniRT, the QuickTalk runtime environment, and model files. After deploying an instance, open port `5173` and visit the instance URL provided by the platform. If you need to restart services manually, follow the commands in the guide.
@@ -162,7 +162,7 @@ source .venv/bin/activate
cp .env.example .env
```
Edit `.env` and configure at least an LLM. The default TTS can use the keyless `edge` voice. LLM, STT, and TTS are independent providers; see [Configuration](docs/en/user-guide/configuration.md) and [LLM / STT](docs/en/model-deployment/llm-stt.md).
Edit `.env` and configure at least an LLM. The default TTS can use the keyless `edge` voice. LLM, STT, and TTS are independent providers; see [Configuration](https://datascale-ai.github.io/opentalking/latest/en/reference/configuration/) and [LLM / STT](https://datascale-ai.github.io/opentalking/latest/en/speech_models/llm-stt/).
```bash
bash scripts/start_unified.sh --mock
@@ -202,12 +202,12 @@ bash scripts/start_unified.sh \
More entrypoints:
- [QuickTalk local deployment](docs/en/model-deployment/quicktalk.md)
- [Wav2Lip local deployment](docs/en/model-deployment/wav2lip-local.md)
- [FasterLivePortrait / JoyVASA](docs/en/model-deployment/fasterliveportrait.md)
- [Video clone guide](docs/en/usage/webui/video-clone.md)
- [WebUI guide](docs/en/usage/webui/basic.md)
- [Docker Compose and production deployment](docs/en/user-guide/deployment.md)
- [QuickTalk local deployment](https://datascale-ai.github.io/opentalking/latest/en/avatar_models/deployment/quicktalk-local/)
- [Wav2Lip local deployment](https://datascale-ai.github.io/opentalking/latest/en/avatar_models/deployment/wav2lip-local/)
- [FasterLivePortrait / JoyVASA](https://datascale-ai.github.io/opentalking/latest/en/avatar_models/fasterliveportrait/)
- [Video clone guide](https://datascale-ai.github.io/opentalking/latest/en/usage/webui/video-clone/)
- [WebUI guide](https://datascale-ai.github.io/opentalking/latest/en/usage/webui/basic/)
- [Docker Compose and production deployment](https://datascale-ai.github.io/opentalking/latest/en/deployment/)
## Supported Models
@@ -227,7 +227,7 @@ More entrypoints:
| --- | --- | --- | --- | --- | --- |
| `quicktalk` | RTX 3090 | Template video + audio | 720x900 / 25fps | About 3.8 GiB | About 35 fps |
For weight downloads, Docker, troubleshooting, and model configuration, see [Model deployment](docs/en/model-deployment/index.md).
For weight downloads, Docker, troubleshooting, and model configuration, see [Model deployment](https://datascale-ai.github.io/opentalking/latest/en/model-deployment/).
### Cloud Model API: Atlas Cloud
@@ -239,7 +239,7 @@ For weight downloads, Docker, troubleshooting, and model configuration, see [Mod
> **[Atlas Cloud](https://www.atlascloud.ai/?utm_source=github&utm_medium=link&utm_campaign=opentalking)** is an all-modal AI inference platform. One API gives you access to video generation, image generation, and LLMs, so you do not need to integrate multiple vendors separately. A single integration can route to 300+ curated all-modal models.
OpenTalking uses an OpenAI-compatible interface for LLMs. Point `OPENTALKING_LLM_BASE_URL` to `https://api.atlascloud.ai/v1` to use Atlas-hosted DeepSeek / Qwen models. See [LLM and STT](docs/en/model-deployment/llm-stt.md). For budget-friendly API options, see Atlas Cloud's [coding plan](https://www.atlascloud.ai/console/coding-plan).
OpenTalking uses an OpenAI-compatible interface for LLMs. Point `OPENTALKING_LLM_BASE_URL` to `https://api.atlascloud.ai/v1` to use Atlas-hosted DeepSeek / Qwen models. See [LLM and STT](https://datascale-ai.github.io/opentalking/latest/en/speech_models/llm-stt/). For budget-friendly API options, see Atlas Cloud's [coding plan](https://www.atlascloud.ai/console/coding-plan).
## Progress And Roadmap
@@ -304,12 +304,12 @@ OpenTalking uses an OpenAI-compatible interface for LLMs. Point `OPENTALKING_LLM
## Documentation And Community
- [Quickstart](docs/en/user-guide/quickstart.md)
- [Models](docs/en/model-deployment/index.md) (weight downloads, mirrors, startup, validation)
- [Architecture](docs/en/developer-guide/architecture.md)
- [Configuration](docs/en/user-guide/configuration.md)
- [Deployment](docs/en/user-guide/deployment.md) (Docker Compose, distributed deployment)
- [Model adapter](docs/en/developer-guide/model-adapter.md)
- [Quickstart](https://datascale-ai.github.io/opentalking/latest/en/quick-start/)
- [Models](https://datascale-ai.github.io/opentalking/latest/en/model-deployment/) (weight downloads, mirrors, startup, validation)
- [Architecture](https://datascale-ai.github.io/opentalking/latest/en/developer-guide/architecture/)
- [Configuration](https://datascale-ai.github.io/opentalking/latest/en/reference/configuration/)
- [Deployment](https://datascale-ai.github.io/opentalking/latest/en/deployment/) (Docker Compose, distributed deployment)
- [Model adapter](https://datascale-ai.github.io/opentalking/latest/en/developer-guide/model-adapter/)
- [Contributing](CONTRIBUTING.md) (dev environment, CLI tools, ruff / mypy / pytest)
Join the QQ community to discuss real-time digital humans, FlashTalk, OmniRT, model deployment, and product scenarios.

44
README.zh.md
View File

@@ -123,12 +123,12 @@ OpenTalking 的 **编排层**（API / Worker / 前端）和 **数字人合成后
| 路线 | 推荐模型 / 后端 | 设备参考 | 适合场景 | 详细文档 |
| --- | --- | --- | --- | --- |
| 快速体验 | `mock` | CPU / 无 GPU | 不下载模型权重，先验证 API、LLM、TTS、WebRTC 与浏览器播放链路 | [快速开始](docs/zh/user-guide/quickstart.md) |
| 入门验证 | `quicktalk` / `wav2lip` | RTX 3050 Laptop、RTX 3060、RTX 4060 | 能跑通真实视频渲染，适合功能演示和部署验证；低显存设备建议降低分辨率 | [QuickTalk](docs/zh/model-deployment/quicktalk.md) / [Wav2Lip](docs/zh/model-deployment/wav2lip-local.md) |
| 消费级显卡单机 | `quicktalk` / `wav2lip` / `musetalk` | RTX 3090、RTX 4090 | 更接近实时体验，适合本地 demo、私有化验证和轻量生产前评估 | [模型部署](docs/zh/model-deployment/index.md) |
| 全本地私有化 | `sensevoice` + `local_cosyvoice` + `quicktalk` | RTX 3090 / 4090 或同级 GPU | STT、TTS、视频驱动都走本地；OpenTalking 使用主 `.venv`，CosyVoice 使用独立 sidecar venv | [本地 STT/TTS + QuickTalk](docs/zh/model-deployment/local-quicktalk-audio.md) |
| 高质量远端推理 | `flashtalk` / `flashhead` / `fasterliveportrait` + OmniRT | 多卡 GPU、Ascend 910B2、远端 GPU 服务 | 多卡、GPU/NPU、生产隔离、更高画质或视频克隆 | [FlashTalk](docs/zh/model-deployment/flashtalk.md) / [FasterLivePortrait](docs/zh/model-deployment/fasterliveportrait.md) |
| Docker / 生产部署 | API、Web、Worker、外部模型服务分离 | 单机 GPU、远端 GPU、分布式集群 | 服务化部署、远端 GPU、分布式和生产验证 | [部署文档](docs/zh/user-guide/deployment.md) |
| 快速体验 | `mock` | CPU / 无 GPU | 不下载模型权重，先验证 API、LLM、TTS、WebRTC 与浏览器播放链路 | [快速开始](https://datascale-ai.github.io/opentalking/latest/quick-start/) |
| 入门验证 | `quicktalk` / `wav2lip` | RTX 3050 Laptop、RTX 3060、RTX 4060 | 能跑通真实视频渲染，适合功能演示和部署验证；低显存设备建议降低分辨率 | [QuickTalk](https://datascale-ai.github.io/opentalking/latest/avatar_models/deployment/quicktalk-local/) / [Wav2Lip](https://datascale-ai.github.io/opentalking/latest/avatar_models/deployment/wav2lip-local/) |
| 消费级显卡单机 | `quicktalk` / `wav2lip` / `musetalk` | RTX 3090、RTX 4090 | 更接近实时体验，适合本地 demo、私有化验证和轻量生产前评估 | [模型与后端选择](https://datascale-ai.github.io/opentalking/latest/model-support/selection/) |
| 全本地私有化 | `sensevoice` + `local_cosyvoice` + `quicktalk` | RTX 3090 / 4090 或同级 GPU | STT、TTS、视频驱动都走本地；OpenTalking 使用主 `.venv`，CosyVoice 使用独立 sidecar venv | [本地 STT/TTS + QuickTalk](https://datascale-ai.github.io/opentalking/latest/recipes/local-quicktalk-audio/) |
| 高质量远端推理 | `flashtalk` / `flashhead` / `fasterliveportrait` + OmniRT | 多卡 GPU、Ascend 910B2、远端 GPU 服务 | 多卡、GPU/NPU、生产隔离、更高画质或视频克隆 | [FlashTalk](https://datascale-ai.github.io/opentalking/latest/avatar_models/flashtalk/) / [FasterLivePortrait](https://datascale-ai.github.io/opentalking/latest/avatar_models/fasterliveportrait/) |
| Docker / 生产部署 | API、Web、Worker、外部模型服务分离 | 单机 GPU、远端 GPU、分布式集群 | 服务化部署、远端 GPU、分布式和生产验证 | [部署文档](https://datascale-ai.github.io/opentalking/latest/deployment/) |
## 快速开始
@@ -145,7 +145,7 @@ OpenTalking 的 **编排层**（API / Worker / 前端）和 **数字人合成后
- 镜像地址：[镜像链接](https://www.compshare.cn/images/TdDwmKZUZebI?referral_code=Hid5KUhcqlZEptmMEwKy2F)
- 对外端口：`5173`（WebUI，内部自动代理 API）
- 操作文档：[优云智算镜像快速体验](docs/zh/quick-start/compshare-image.md)
- 操作文档：[优云智算镜像快速体验](https://datascale-ai.github.io/opentalking/latest/quick-start/)
镜像内已预置 OpenTalking、OmniRT、QuickTalk 运行环境和模型文件。部署实例后开放 `5173` 端口，在浏览器访问平台提供的实例地址即可进入 WebUI；如需手动重启服务，请按操作文档中的命令执行。
@@ -162,7 +162,7 @@ source .venv/bin/activate
cp .env.example .env
```
编辑 `.env`，至少配置 LLM；TTS 默认可使用不需要 key 的 `edge` 语音。LLM、STT、TTS 是独立 provider，常见配置见 [配置说明](docs/zh/user-guide/configuration.md) 和 [LLM / STT 文档](docs/zh/model-deployment/llm-stt.md)。
编辑 `.env`，至少配置 LLM；TTS 默认可使用不需要 key 的 `edge` 语音。LLM、STT、TTS 是独立 provider，常见配置见 [配置说明](https://datascale-ai.github.io/opentalking/latest/reference/configuration/) 和 [LLM / STT 文档](https://datascale-ai.github.io/opentalking/latest/speech_models/llm-stt/)。
```bash
bash scripts/start_unified.sh --mock
@@ -202,12 +202,12 @@ bash scripts/start_unified.sh \
更多入口：
- [QuickTalk 本地部署](docs/zh/model-deployment/quicktalk.md)
- [Wav2Lip 本地部署](docs/zh/model-deployment/wav2lip-local.md)
- [FasterLivePortrait / JoyVASA](docs/zh/model-deployment/fasterliveportrait.md)
- [视频克隆使用说明](docs/zh/usage/webui/video-clone.md)
- [WebUI 使用说明](docs/zh/usage/webui/basic.md)
- [Docker Compose 与生产部署](docs/zh/user-guide/deployment.md)
- [QuickTalk 本地部署](https://datascale-ai.github.io/opentalking/latest/avatar_models/deployment/quicktalk-local/)
- [Wav2Lip 本地部署](https://datascale-ai.github.io/opentalking/latest/avatar_models/deployment/wav2lip-local/)
- [FasterLivePortrait / JoyVASA](https://datascale-ai.github.io/opentalking/latest/avatar_models/fasterliveportrait/)
- [视频克隆使用说明](https://datascale-ai.github.io/opentalking/latest/usage/webui/video-clone/)
- [WebUI 使用说明](https://datascale-ai.github.io/opentalking/latest/usage/webui/basic/)
- [Docker Compose 与生产部署](https://datascale-ai.github.io/opentalking/latest/deployment/)
## 模型支持
@@ -227,7 +227,7 @@ bash scripts/start_unified.sh \
| --- | --- | --- | --- | --- | --- |
| `quicktalk` | RTX 3090 | template video + audio | 720x900 / 25fps | 约 3.8 GiB | 约 35 fps |
更多权重下载、Docker、故障排查和模型配置见 [模型部署索引](docs/zh/model-deployment/index.md)。
更多权重下载、Docker、故障排查和模型配置见 [模型部署索引](https://datascale-ai.github.io/opentalking/latest/model-deployment/)。
### 云端模型 API：Atlas Cloud
@@ -239,7 +239,7 @@ bash scripts/start_unified.sh \
> 🎁 **[Atlas Cloud](https://www.atlascloud.ai/?utm_source=github&utm_medium=link&utm_campaign=opentalking)** 是一个全模态 AI 推理平台，用一套 API 即可访问视频生成、图像生成和 LLM，无需分别对接多家厂商，一次接入即可统一调用全模态的 300+ 精选模型。
OpenTalking 的 LLM 走 OpenAI-compatible 接口，把 `OPENTALKING_LLM_BASE_URL` 指向 `https://api.atlascloud.ai/v1` 即可直接使用 Atlas 托管的 DeepSeek / Qwen 等模型，配置见 [LLM 与 STT](docs/zh/model-deployment/llm-stt.md)。更多预算友好的 API 方案见 Atlas Cloud [coding plan](https://www.atlascloud.ai/console/coding-plan)。
OpenTalking 的 LLM 走 OpenAI-compatible 接口，把 `OPENTALKING_LLM_BASE_URL` 指向 `https://api.atlascloud.ai/v1` 即可直接使用 Atlas 托管的 DeepSeek / Qwen 等模型，配置见 [LLM 与 STT](https://datascale-ai.github.io/opentalking/latest/speech_models/llm-stt/)。更多预算友好的 API 方案见 Atlas Cloud [coding plan](https://www.atlascloud.ai/console/coding-plan)。
## 能力进展与 Roadmap
@@ -302,12 +302,12 @@ OpenTalking 的 LLM 走 OpenAI-compatible 接口，把 `OPENTALKING_LLM_BASE_URL
## 文档与社区
- [快速开始](docs/zh/user-guide/quickstart.md)
- [模型](docs/zh/model-deployment/index.md)（权重下载、国内源、启动、验证）
- [架构说明](docs/zh/developer-guide/architecture.md)
- [配置说明](docs/zh/user-guide/configuration.md)
- [部署文档](docs/zh/user-guide/deployment.md)（Docker Compose、分布式部署）
- [模型适配](docs/zh/developer-guide/model-adapter.md)
- [快速开始](https://datascale-ai.github.io/opentalking/latest/quick-start/)
- [模型](https://datascale-ai.github.io/opentalking/latest/model-deployment/)（权重下载、国内源、启动、验证）
- [架构说明](https://datascale-ai.github.io/opentalking/latest/developer-guide/architecture/)
- [配置说明](https://datascale-ai.github.io/opentalking/latest/reference/configuration/)
- [部署文档](https://datascale-ai.github.io/opentalking/latest/deployment/)（Docker Compose、分布式部署）
- [模型适配](https://datascale-ai.github.io/opentalking/latest/developer-guide/model-adapter/)
- [贡献指南](CONTRIBUTING.md)（开发环境、CLI 工具、ruff / mypy / pytest）
欢迎加入 QQ 交流群，讨论实时数字人、FlashTalk、OmniRT、模型部署和产品场景。