Realtime-Venus local preparation commited on
Commit
7b362f0
·
1 Parent(s): cbf2133

Refine public model documentation

Browse files
Files changed (2) hide show
  1. README.md +108 -38
  2. README_zh.md +108 -38
README.md CHANGED
@@ -131,11 +131,35 @@ new output directory when repeating an experiment.
131
 
132
  ## 6. 🛠️ Installation
133
 
134
- The models use a shared root configuration and separate model directories. Root `config.yaml` contains the complete Audio and Omni configurations; each model directory contains weights, custom code, the tokenizer, processor, and supporting resources. The unified downloader fetches both the root configuration and the complete selected model directories.
135
 
136
- ### 6.1 Create an environment
137
 
138
- Inference requires an NVIDIA CUDA GPU. The following commands create a Python 3.10 environment with Conda; use a compatible PyTorch and CUDA environment.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
139
 
140
  ```bash
141
  conda create -n realtime-venus python=3.10 -y
@@ -147,9 +171,9 @@ mkdir -p Realtime-Venus
147
  cd Realtime-Venus
148
  ```
149
 
150
- ### 6.2 Download the models
151
 
152
- Fetch the unified downloader first. The commands below use one fixed model revision:
153
 
154
  ```bash
155
  MODEL_REVISION=03b5c2919dd684900bdf6852272ac6a8f30fab74
@@ -157,46 +181,32 @@ hf download inclusionAI/Realtime-Venus download_models.py \
157
  --revision "$MODEL_REVISION" --local-dir .
158
  ```
159
 
160
- Choose the value of `--model`:
161
 
162
- | Value | Downloads |
163
- | --- | --- |
164
- | `omni` | Realtime-Venus-Omni for audio-visual understanding and interaction |
165
- | `audio` | Realtime-Venus-Audio for audio understanding and conversation |
166
- | `all` | Both models |
167
 
168
- For example, download both models:
 
169
 
170
- ```bash
171
  python download_models.py --model all --local-dir . --revision "$MODEL_REVISION"
172
  ```
173
 
174
- To download one model, replace `all` with `omni` or `audio`. The downloader reads root `config.yaml` first and then downloads the complete selected model directories. It does not download `data/HEARTBEAT_Challenge/`. If interrupted, rerun the same command to reuse completed files.
175
-
176
- The main files after downloading both models are shown below. Selecting one model creates only its corresponding model directory:
177
 
178
- ```text
179
- Realtime-Venus/
180
- ├── config.yaml
181
- ├── download_models.py
182
- ├── Realtime-Venus-Omni/
183
- │ ├── config.json
184
- │ ├── requirements.txt
185
- │ └── … # Weights, custom code, tokenizer, processor, and resources
186
- └── Realtime-Venus-Audio/
187
- ├── config.json
188
- └── … # Weights, custom code, tokenizer, processor, and resources
189
- ```
190
 
191
- ### 6.3 Install inference dependencies
192
 
193
- Both models share `Realtime-Venus-Omni/requirements.txt`. After downloading Omni or `all`, run:
194
 
195
  ```bash
196
  python -m pip install -r Realtime-Venus-Omni/requirements.txt
197
  ```
198
 
199
- For an Audio-only download, fetch this dependency file separately before installing. This adds only the Omni `requirements.txt` file:
200
 
201
  ```bash
202
  hf download inclusionAI/Realtime-Venus Realtime-Venus-Omni/requirements.txt \
@@ -204,23 +214,83 @@ hf download inclusionAI/Realtime-Venus Realtime-Venus-Omni/requirements.txt \
204
  python -m pip install -r Realtime-Venus-Omni/requirements.txt
205
  ```
206
 
207
- ### 6.4 Use the local model directories
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
208
 
209
- Run the inference examples in Sections 7 and 8 from the `Realtime-Venus/` working directory above. The model paths are `./Realtime-Venus-Omni` and `./Realtime-Venus-Audio`. For example, read the local configuration first:
 
 
210
 
211
  ```python
212
- from transformers import AutoConfig
 
 
213
 
214
- config = AutoConfig.from_pretrained(
215
- "./Realtime-Venus-Omni", # Use "./Realtime-Venus-Audio" for Audio
 
 
 
 
 
 
 
216
  trust_remote_code=True,
217
  local_files_only=True,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
218
  )
 
 
 
 
 
219
  ```
220
 
221
- Root `config.yaml` works together with each model directory's `config.json`. When moving the models or deploying offline, retain root `config.yaml` alongside the complete model directories and install dependencies in advance. Preserve the relative directory layout above so configuration can be read locally. Load the selected model subdirectory for inference; asset and output paths are relative to the working directory.
 
 
 
 
222
 
223
- The model revision is `03b5c2919dd684900bdf6852272ac6a8f30fab74`. Each model's `venus_config_ref` pins root configuration revision `15d5b420cd8557e9582d41a57a80af3e4de18e47`. The download commands pin the model revision, and the model reference verifies the root configuration's SHA-256 digest.
224
 
225
  ## 7. 🎙️ Realtime-Venus-Omni Usages
226
 
 
131
 
132
  ## 6. 🛠️ Installation
133
 
134
+ ### 6.1 Use the new repository layout
135
 
136
+ Use the repository root as your local working directory, keeping the root configuration alongside the complete selected model directories. Run the commands and examples below from that working directory.
137
 
138
+ ```text
139
+ Realtime-Venus/
140
+ ├── config.yaml # Complete model configurations and directory mapping
141
+ ├── download_models.py # Unified downloader
142
+ ├── Realtime-Venus-Omni/ # Audio-visual model; download as needed
143
+ │ ├── config.json
144
+ │ ├── requirements.txt
145
+ │ ├── assets/
146
+ │ └── … # Weights, custom code, tokenizer, processor, etc.
147
+ └── Realtime-Venus-Audio/ # Audio-only model; download as needed
148
+ ├── config.json
149
+ ├── assets/
150
+ └── … # Weights, custom code, tokenizer, processor, etc.
151
+ ```
152
+
153
+ | File or directory | How to use it |
154
+ | --- | --- |
155
+ | Root `config.yaml` | Contains the complete model configurations; the downloader fetches it automatically and the custom configuration code reads it when loading a model |
156
+ | Each model's `config.json` | Identifies the model to Transformers and references a fixed root configuration revision |
157
+ | `Realtime-Venus-Omni/` | Load this subdirectory for audio-visual understanding, conversation, and full-duplex interaction |
158
+ | `Realtime-Venus-Audio/` | Load this subdirectory for audio understanding, conversation, and full-duplex interaction |
159
+
160
+ ### 6.2 Prepare the environment
161
+
162
+ Inference requires an NVIDIA CUDA GPU. The following commands use Python 3.10, Conda, and FFmpeg. Use compatible versions of PyTorch, CUDA, and your GPU driver.
163
 
164
  ```bash
165
  conda create -n realtime-venus python=3.10 -y
 
171
  cd Realtime-Venus
172
  ```
173
 
174
+ ### 6.3 Download the models you need
175
 
176
+ Fetch the unified downloader and set the model revision used in this section. Run the following commands in the same terminal:
177
 
178
  ```bash
179
  MODEL_REVISION=03b5c2919dd684900bdf6852272ac6a8f30fab74
 
181
  --revision "$MODEL_REVISION" --local-dir .
182
  ```
183
 
184
+ Choose one of the following three commands:
185
 
186
+ ```bash
187
+ # Omni only: audio-visual interaction
188
+ python download_models.py --model omni --local-dir . --revision "$MODEL_REVISION"
 
 
189
 
190
+ # Audio only: audio interaction
191
+ python download_models.py --model audio --local-dir . --revision "$MODEL_REVISION"
192
 
193
+ # Both models
194
  python download_models.py --model all --local-dir . --revision "$MODEL_REVISION"
195
  ```
196
 
197
+ The downloader fetches root `config.yaml` first, then uses its directory mapping to download the complete selected model directories, including weights, custom code, tokenizer, processor, TTS resources, and demo assets. Selecting one model does not download the other model's weights. These commands also do not download `data/HEARTBEAT_Challenge/`.
 
 
198
 
199
+ If interrupted, rerun the same download command to reuse completed files. Change `--local-dir` to use a different destination. When running examples, enter that directory or adjust the model and asset paths accordingly.
 
 
 
 
 
 
 
 
 
 
 
200
 
201
+ ### 6.4 Install inference dependencies
202
 
203
+ Both models share `Realtime-Venus-Omni/requirements.txt`. After downloading Omni or both models, install it directly:
204
 
205
  ```bash
206
  python -m pip install -r Realtime-Venus-Omni/requirements.txt
207
  ```
208
 
209
+ For an Audio-only download, fetch this dependency file before installing. This adds only the Omni `requirements.txt` file and does not download Omni weights:
210
 
211
  ```bash
212
  hf download inclusionAI/Realtime-Venus Realtime-Venus-Omni/requirements.txt \
 
214
  python -m pip install -r Realtime-Venus-Omni/requirements.txt
215
  ```
216
 
217
+ ### 6.5 Load and use the models
218
+
219
+ #### Omni: load the local model subdirectory
220
+
221
+ From the `Realtime-Venus/` working directory, use Transformers to load `./Realtime-Venus-Omni`:
222
+
223
+ ```python
224
+ import torch
225
+ from transformers import AutoModel
226
+
227
+ model = AutoModel.from_pretrained(
228
+ "./Realtime-Venus-Omni",
229
+ trust_remote_code=True,
230
+ local_files_only=True,
231
+ attn_implementation="sdpa",
232
+ torch_dtype=torch.bfloat16,
233
+ ).eval().cuda()
234
+ ```
235
+
236
+ After loading, follow Section 7.2 for full-duplex video interaction or Section 7.3 for offline conversation over a complete video. Full-duplex interaction uses `as_duplex()`, `prepare()`, `streaming_prefill()`, and `streaming_generate()`; offline conversation uses `model.chat(...)`. Follow the corresponding example to read input files from `Realtime-Venus-Omni/assets/`.
237
 
238
+ #### Audio: run one audio question-answering example
239
+
240
+ Save the following code as `run_audio.py` in the same working directory. It loads the local Audio model, reads its bundled demo audio, and prints a text response:
241
 
242
  ```python
243
+ import librosa
244
+ import torch
245
+ from transformers import AutoModel, AutoTokenizer
246
 
247
+ model_dir = "./Realtime-Venus-Audio"
248
+ tokenizer = AutoTokenizer.from_pretrained(
249
+ model_dir,
250
+ trust_remote_code=True,
251
+ local_files_only=True,
252
+ fix_mistral_regex=True,
253
+ )
254
+ model = AutoModel.from_pretrained(
255
+ model_dir,
256
  trust_remote_code=True,
257
  local_files_only=True,
258
+ attn_implementation="sdpa",
259
+ torch_dtype=torch.bfloat16,
260
+ init_vision=False,
261
+ init_audio=True,
262
+ init_tts=False,
263
+ ).eval().cuda()
264
+
265
+ audio, _ = librosa.load(
266
+ "Realtime-Venus-Audio/assets/case_offline.wav", sr=16000, mono=True
267
+ )
268
+ answer = model.chat(
269
+ msgs=[
270
+ {"role": "system", "content": "You are a helpful assistant."},
271
+ {"role": "user", "content": [audio, "What is the speaker asking about?"]},
272
+ ],
273
+ tokenizer=tokenizer,
274
+ do_sample=False,
275
+ max_new_tokens=2048,
276
+ enable_thinking=False,
277
+ use_tts_template=True,
278
+ generate_audio=False,
279
  )
280
+ print(answer)
281
+ ```
282
+
283
+ ```bash
284
+ python run_audio.py
285
  ```
286
 
287
+ This example outputs text only and therefore sets `init_tts=False`. For speech output or full-duplex interaction, enable `init_tts=True` and follow the loading and calling examples in Section 8.
288
+
289
+ ### 6.6 Offline use and moving the models
290
+
291
+ After downloading the models and installing dependencies while online, load them using the local paths above. When moving to another machine, copy root `config.yaml` together with the complete model directories you need. Preserve the relative layout in Section 6.1 and install the dependencies on the target machine. The model option `local_files_only=True` uses local files.
292
 
293
+ Each model's `config.json` references root configuration revision `15d5b420cd8557e9582d41a57a80af3e4de18e47` through `venus_config_ref` and verifies its SHA-256 digest. The unified downloader fetches the corresponding root configuration. Point `from_pretrained` at the selected model subdirectory, such as `./Realtime-Venus-Audio`.
294
 
295
  ## 7. 🎙️ Realtime-Venus-Omni Usages
296
 
README_zh.md CHANGED
@@ -91,11 +91,35 @@
91
 
92
  ## 6. 🛠️ 安装
93
 
94
- 模型按“根配置 + 独立模型目录”组织:根目录的 `config.yaml` 保存 Audio 和 Omni 的完整模型配置;各模型目录包含权重、自定义代码、Tokenizer、Processor 和配套资源。统一下载脚本会一并下载根配置与所选模型的完整目录。
95
 
96
- ### 6.1 创建环境
97
 
98
- 推理需要 NVIDIA CUDA GPU。以下使用 Python 3.10 和 Conda 创建环境;PyTorch 与 CUDA 环境需兼容。
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
99
 
100
  ```bash
101
  conda create -n realtime-venus python=3.10 -y
@@ -107,9 +131,9 @@ mkdir -p Realtime-Venus
107
  cd Realtime-Venus
108
  ```
109
 
110
- ### 6.2 下载模型
111
 
112
- 先获取统一下载脚本。以下命令使用同一个固定模型版本:
113
 
114
  ```bash
115
  MODEL_REVISION=03b5c2919dd684900bdf6852272ac6a8f30fab74
@@ -117,46 +141,32 @@ hf download inclusionAI/Realtime-Venus download_models.py \
117
  --revision "$MODEL_REVISION" --local-dir .
118
  ```
119
 
120
- 根据需要选择 `--model`:
121
 
122
- | 参数 | 下载内容 |
123
- | --- | --- |
124
- | `omni` | Realtime-Venus-Omni:音视频理解与交互 |
125
- | `audio` | Realtime-Venus-Audio:音频理解与语音对话 |
126
- | `all` | 两款模型 |
127
 
128
- 例如,下载两款模型:
 
129
 
130
- ```bash
131
  python download_models.py --model all --local-dir . --revision "$MODEL_REVISION"
132
  ```
133
 
134
- 只需一款时,将上面的 `all` 改为 `omni` 或 `audio`。下载器先读取根 `config.yaml`,再下载选定模型的完整目录,不会下载 `data/HEARTBEAT_Challenge/`。中断后重新执行同一条命令即可复用已经下载完成的文件。
135
-
136
- 下载两款模型后的主要目录如下;只下载一款时,仅有对应的模型目录:
137
 
138
- ```text
139
- Realtime-Venus/
140
- ├── config.yaml
141
- ├── download_models.py
142
- ├── Realtime-Venus-Omni/
143
- │ ├── config.json
144
- │ ├── requirements.txt
145
- │ └── … # 权重、自定义代码、Tokenizer、Processor 与资源
146
- └── Realtime-Venus-Audio/
147
- ├── config.json
148
- └── … # 权重、自定义代码、Tokenizer、Processor 与资源
149
- ```
150
 
151
- ### 6.3 安装推理依赖
152
 
153
- 两款模型共用 `Realtime-Venus-Omni/requirements.txt`。下载了 Omni 或 `all` 时,直接运行:
154
 
155
  ```bash
156
  python -m pip install -r Realtime-Venus-Omni/requirements.txt
157
  ```
158
 
159
- 如果只下载了 Audio,先单独获取这份依赖文件,再安装;这一步只会增加 Omni 的 `requirements.txt` 文件:
160
 
161
  ```bash
162
  hf download inclusionAI/Realtime-Venus Realtime-Venus-Omni/requirements.txt \
@@ -164,23 +174,83 @@ hf download inclusionAI/Realtime-Venus Realtime-Venus-Omni/requirements.txt \
164
  python -m pip install -r Realtime-Venus-Omni/requirements.txt
165
  ```
166
 
167
- ### 6.4 从本地目录使用
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
168
 
169
- 在上述 `Realtime-Venus/` 工作目录中运行第 7、8 节的推理示例,模型路径分别为 `./Realtime-Venus-Omni` 和 `./Realtime-Venus-Audio`。例如,可先读取本地配置:
 
 
170
 
171
  ```python
172
- from transformers import AutoConfig
 
 
173
 
174
- config = AutoConfig.from_pretrained(
175
- "./Realtime-Venus-Omni", # Audio 改为 "./Realtime-Venus-Audio"
 
 
 
 
 
 
 
176
  trust_remote_code=True,
177
  local_files_only=True,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
178
  )
 
 
 
 
 
179
  ```
180
 
181
- 根 `config.yaml` 与模型子目录中的 `config.json` 配合使用。搬移模型或离线部署时,请一起保留根 `config.yaml` 和完整的模型目录,并提前安装依赖;保持上面的相对目录关系即可从本地读取配置。推理时加载对应模型子目录,资源路径与输出路径均相对于工作目录。
 
 
 
 
182
 
183
- 模型版本为 `03b5c2919dd684900bdf6852272ac6a8f30fab74`;子模型配置通过 `venus_config_ref` 引用根配置版本 `15d5b420cd8557e9582d41a57a80af3e4de18e47`。下载命令已固定模型版本,根配置内容由子模型引用中的 SHA-256 校验。
184
 
185
  ## 7. 🎙️ Realtime-Venus-Omni 使用方法
186
 
 
91
 
92
  ## 6. 🛠️ 安装
93
 
94
+ ### 6.1 新版仓库的使用方式
95
 
96
+ 下载模型时,以仓库根目录为本地工作目录,保留根配置和所选模型的完整子目录。后续所有命令和示例均在这个工作目录中执行。
97
 
98
+ ```text
99
+ Realtime-Venus/
100
+ ├── config.yaml # 两款模型的完整配置及目录映射
101
+ ├── download_models.py # 统一下载入口
102
+ ├── Realtime-Venus-Omni/ # 音视频模型,按需下载
103
+ │ ├── config.json
104
+ │ ├── requirements.txt
105
+ │ ├── assets/
106
+ │ └── … # 权重、自定义代码、Tokenizer、Processor 等
107
+ └── Realtime-Venus-Audio/ # 纯音频模型,按需下载
108
+ ├── config.json
109
+ ├── assets/
110
+ └── … # 权重、自定义代码、Tokenizer、Processor 等
111
+ ```
112
+
113
+ | 文件或目录 | 如何使用 |
114
+ | --- | --- |
115
+ | 根 `config.yaml` | 保存完整模型配置;统一下载器会自动获取,加载模型时由自定义配置代码读取 |
116
+ | 各模型目录的 `config.json` | 供 Transformers 识别模型类型,并引用固定版本的根配置 |
117
+ | `Realtime-Venus-Omni/` | 音视频理解、对话和全双工交互,从这个子目录加载模型 |
118
+ | `Realtime-Venus-Audio/` | 音频理解、语音对话和全双工交互,从这个子目录加载模型 |
119
+
120
+ ### 6.2 准备环境
121
+
122
+ 推理需要 NVIDIA CUDA GPU。以下使用 Python 3.10、Conda 和 FFmpeg;请使用相互兼容的 PyTorch、CUDA 和显卡驱动。
123
 
124
  ```bash
125
  conda create -n realtime-venus python=3.10 -y
 
131
  cd Realtime-Venus
132
  ```
133
 
134
+ ### 6.3 按需下载模型
135
 
136
+ 先下载统一下载脚本,并指定本节使用的模型版本。以下命令应在同一个终端中执行:
137
 
138
  ```bash
139
  MODEL_REVISION=03b5c2919dd684900bdf6852272ac6a8f30fab74
 
141
  --revision "$MODEL_REVISION" --local-dir .
142
  ```
143
 
144
+ 根据需要,从下面三条命令中选择一条执行:
145
 
146
+ ```bash
147
+ # 只下载 Omni:用于音视频交互
148
+ python download_models.py --model omni --local-dir . --revision "$MODEL_REVISION"
 
 
149
 
150
+ # 只下载 Audio:用于纯音频交互
151
+ python download_models.py --model audio --local-dir . --revision "$MODEL_REVISION"
152
 
153
+ # 同时下载两款模型
154
  python download_models.py --model all --local-dir . --revision "$MODEL_REVISION"
155
  ```
156
 
157
+ 下载器会先下载根 `config.yaml`,再根据其中的目录映射下载所选模型的完整目录,包括权重、自定义代码、Tokenizer、Processor、TTS 和演示资源。选择一款模型时,不会下载另一款模型的权重;以上命令也不会下载 `data/HEARTBEAT_Challenge/` 数据。
 
 
158
 
159
+ 中断后,重新执行相同的下载命令即可复用已完成的文件。若需要更换本地保存位置,可修改 `--local-dir`;运行示例时,进入该目录,或相应调整模型和资源文件路径。
 
 
 
 
 
 
 
 
 
 
 
160
 
161
+ ### 6.4 安装推理依赖
162
 
163
+ 两款模型共用 `Realtime-Venus-Omni/requirements.txt`。如果下载了 Omni 或两款模型,直接安装:
164
 
165
  ```bash
166
  python -m pip install -r Realtime-Venus-Omni/requirements.txt
167
  ```
168
 
169
+ 如果只下载了 Audio,先获取这一份依赖文件,再安装。该命令只会增加 Omni 目录下的 `requirements.txt`,不会下载 Omni 权重:
170
 
171
  ```bash
172
  hf download inclusionAI/Realtime-Venus Realtime-Venus-Omni/requirements.txt \
 
174
  python -m pip install -r Realtime-Venus-Omni/requirements.txt
175
  ```
176
 
177
+ ### 6.5 加载并使用模型
178
+
179
+ #### Omni:从本地子目录加载
180
+
181
+ 在 `Realtime-Venus/` 工作目录中,使用 Transformers 加载 `./Realtime-Venus-Omni`:
182
+
183
+ ```python
184
+ import torch
185
+ from transformers import AutoModel
186
+
187
+ model = AutoModel.from_pretrained(
188
+ "./Realtime-Venus-Omni",
189
+ trust_remote_code=True,
190
+ local_files_only=True,
191
+ attn_implementation="sdpa",
192
+ torch_dtype=torch.bfloat16,
193
+ ).eval().cuda()
194
+ ```
195
+
196
+ 加载后,第 7.2 节给出全双工视频交互示例,第 7.3 节给出完整视频的离线对话示例。全双工使用 `as_duplex()`、`prepare()`、`streaming_prefill()` 和 `streaming_generate()`;离线对话使用 `model.chat(...)`。按对应示例读取 `Realtime-Venus-Omni/assets/` 中的输入文件。
197
 
198
+ #### Audio:运行一次音频问答
199
+
200
+ 在同一工作目录中,将下面代码保存为 `run_audio.py`。它加载本地 Audio 模型,读取随模型下载的演示音频,并打印文本回答:
201
 
202
  ```python
203
+ import librosa
204
+ import torch
205
+ from transformers import AutoModel, AutoTokenizer
206
 
207
+ model_dir = "./Realtime-Venus-Audio"
208
+ tokenizer = AutoTokenizer.from_pretrained(
209
+ model_dir,
210
+ trust_remote_code=True,
211
+ local_files_only=True,
212
+ fix_mistral_regex=True,
213
+ )
214
+ model = AutoModel.from_pretrained(
215
+ model_dir,
216
  trust_remote_code=True,
217
  local_files_only=True,
218
+ attn_implementation="sdpa",
219
+ torch_dtype=torch.bfloat16,
220
+ init_vision=False,
221
+ init_audio=True,
222
+ init_tts=False,
223
+ ).eval().cuda()
224
+
225
+ audio, _ = librosa.load(
226
+ "Realtime-Venus-Audio/assets/case_offline.wav", sr=16000, mono=True
227
+ )
228
+ answer = model.chat(
229
+ msgs=[
230
+ {"role": "system", "content": "You are a helpful assistant."},
231
+ {"role": "user", "content": [audio, "What is the speaker asking about?"]},
232
+ ],
233
+ tokenizer=tokenizer,
234
+ do_sample=False,
235
+ max_new_tokens=2048,
236
+ enable_thinking=False,
237
+ use_tts_template=True,
238
+ generate_audio=False,
239
  )
240
+ print(answer)
241
+ ```
242
+
243
+ ```bash
244
+ python run_audio.py
245
  ```
246
 
247
+ 该示例只输出文本,因此设置 `init_tts=False`。需要语音输出或全双工交互时,按照第 8 节的加载和调用示例启用 `init_tts=True`。
248
+
249
+ ### 6.6 离线使用与目录搬移
250
+
251
+ 在联网环境中完成模型下载和依赖安装后,可以按上述本地路径加载。复制到其他机器时,将根 `config.yaml` 与所需的完整模型目录一起复制,保持 6.1 中的相对位置,并为目标机器安装对应依赖。模型参数中的 `local_files_only=True` 使用本地文件。
252
 
253
+ 各模型目录的 `config.json` 通过 `venus_config_ref` 引用根配置版本 `15d5b420cd8557e9582d41a57a80af3e4de18e47`,并校验其 SHA-256;统一下载器已获取对应的根配置。使用模型时,将 `from_pretrained` 指向模型子目录,例如 `./Realtime-Venus-Audio`。
254
 
255
  ## 7. 🎙️ Realtime-Venus-Omni 使用方法
256