diff --git a/README.md b/README.md
index 7ead9abb61ee3cfa6f2230015f7646d4375a312a..f3a782e9515f884b999af397b9bd836cbe5a5499 100644
--- a/README.md
+++ b/README.md
@@ -84,7 +84,6 @@ Remove `#` of the line in `docker-compose.yml`
```
* Config files for self-hosting the FoloToy Server.
-* Config files for self-hosting the FoloToy Server.
# folotoy-server-self-hosting
```
@@ -96,8 +95,6 @@ Remove `#` of the line in `docker-compose.yml`
```
* Config files for self-hosting the FoloToy Server.
-* Config files for self-hosting the FoloToy Server.
-* Config files for self-hosting the FoloToy Server.
* Recommended using **Linux x86_64**, Debian 10-11/Ubuntu 22.04
ELEVENLABS_TTS_MODEL: eleven_multilingual_v2
```
@@ -108,10 +105,18 @@ ### Using Custom Prompt and Voice
**If you use Azure OpenAI, model field must be deployment name you set when deploy models**
+#### Voice List
- [Azure Voice List](https://learn.microsoft.com/zh-cn/azure/ai-services/speech-service/language-support?tabs=tts)
- [Edge-tts Voice List](https://github.com/rany2/edge-tts#changing-the-voice)
+
+#### Language Codes
+- **Required** `OpenAI key` with `whisper-1` and `gpt-3.5-turbo/gpt-4`
* Recommended using **Linux x86_64**, Debian 10-11/Ubuntu 22.04
+- [Whisper language 639-1 codes in roles.json](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes)
+#### Role Level Config Examples
+
+if `stt_type` is `openai-whisper`, `language` should use 639-1 codes
```json
{"1": {
@@ -125,13 +130,75 @@ "top_p": 0.95,
"frequency_penalty": 0,
"presence_penalty": 0,
"voice_name": "zh-CN-XiaoshuangNeural",
+ "language": "zh",
+ "stt_type": "openai-whisper"
## Preparation
+## Preparation
+[English](./README.md) | [简体中文](./README.zh_CN.md)
* Recommended using **Linux x86_64**, Debian 10-11/Ubuntu 22.04
+
+if `stt_type` is `azure-sst`, `language` should use BCP-47 codes
+
+```json
+{"1": {
+ "model": "gpt-3.5-turbo",
+* Recommended using **Linux x86_64**, Debian 10-11/Ubuntu 22.04
## Preparation
+ "prompt": "你是一个知识渊博，乐于助人的智能机器人,你的名字叫“火火兔”，你的任务是陪我聊天，请用简短的对话方式，用中文讲一段话，每次回答不超过50个字！",
+ "max_message_count": 20,
+ "temperature": 0.7,
## Preparation
+ "top_p": 0.95,
+ "frequency_penalty": 0,
+## Preparation
[English](./README.md) | [简体中文](./README.zh_CN.md)
+ "voice_name": "zh-CN-XiaoshuangNeural",
+ "language": "zh-CN",
+ "stt_type": "azure-stt"
+ }}
+
+- or `Azure OpenAI Key` and `Azure Whisper Key`
* Recommended using **Linux x86_64**, Debian 10-11/Ubuntu 22.04
+ ```json
+ {"1": {
+ "model": "gpt-3.5-turbo",
+ "start_text": "你好，我是火火兔，请问有什么我可以帮助你的吗？",
+ "prompt": "你是一个知识渊博，乐于助人的智能机器人,你的名字叫“火火兔”，你的任务是陪我聊天，请用简短的对话方式，用中文讲一段话，每次回答不超过50个字！",
+ "max_message_count": 10,
+ "temperature": 0.7,
+ "max_tokens": 800,
+ "top_p": 0.95,
+ "frequency_penalty": 0,
+ "presence_penalty": 0,
+ "voice_name": "zh-CN-XiaoshuangNeural",
+ "language": "zh-CN",
+ "stt_type": "azure-stt",
+ "llm_type": "openai",
+ "tts_type": "azure-tts"
+ }}
+ ```
+
+ ```json
+ {"1": {
+ "model": "gpt-3.5-turbo",
+ "start_text": "你好，我是火火兔，请问有什么我可以帮助你的吗？",
+ "prompt": "你是一个知识渊博，乐于助人的智能机器人,你的名字叫“火火兔”，你的任务是陪我聊天，请用简短的对话方式，用中文讲一段话，每次回答不超过50个字！",
+ "max_message_count": 10,
+ "temperature": 0.7,
+ "max_tokens": 800,
+ "top_p": 0.95,
+ "frequency_penalty": 0,
+ "presence_penalty": 0,
+ "voice_name": "6xTjFMIxZYYFJag51KZe",
+ "language": "zh-CN",
+ "stt_type": "azure-stt",
+ "llm_type": "azure-openai",
+ "tts_type": "elevenlabs"
+ }}
+ ```
+
+```
### MQTT Authentication
**The default configuration of EMQX allows any anonymous client to access. You can make your EMQX service only allow connections from your own devices by following these steps.**
diff --git a/roles.json b/roles.json
index 2dc9ed3da4547c6617209535baf0441fd9fe4d20..374fd2b8a547d57f3740d8e44cd292a0e8ae0b89 100644
--- a/roles.json
+++ b/roles.json
@@ -75,8 +75,9 @@ "top_p": 0.95,
"frequency_penalty": 0,
"presence_penalty": 0,
"voice_name": "zh-TW-HsiaoChenNeural",
-{
+ "start_text": "你好，我是小兔兔，请问有什么我可以帮助你的吗？",
"model": "gpt-3.5-turbo",
+ "stt_type": "azure-stt"
},
"7": {
"model": "gpt-3.5-turbo",
