修改README.md对openai兼容tts的案例支持

This commit is contained in:
zzh
2025-07-15 18:39:24 +09:00
parent a6aecb5d89
commit 7a7caef1a6
+24 -29
View File
@@ -69,9 +69,9 @@ if "tts" in model_name.lower():
## 📝 使用示例 ## 📝 使用示例
### 单人语音TTS请求(自动启用增强服务) ### 1. 原生Gemini单人TTS请求(使用TTS增强服务)
包含 `voiceConfig.prebuiltVoiceConfig` 的请求会自动使用TTS增强服务: 包含 `voiceConfig.prebuiltVoiceConfig`原生Gemini格式请求会自动使用TTS增强服务:
```bash ```bash
curl -X POST "https://your-domain.com/v1beta/models/gemini-2.5-flash-preview-tts:generateContent" \ curl -X POST "https://your-domain.com/v1beta/models/gemini-2.5-flash-preview-tts:generateContent" \
@@ -96,9 +96,9 @@ curl -X POST "https://your-domain.com/v1beta/models/gemini-2.5-flash-preview-tts
}' }'
``` ```
### 多人语音TTS请求(自动启用增强服务) ### 2. 原生Gemini多人TTS请求(使用TTS增强服务)
包含 `multiSpeakerVoiceConfig` 的请求会自动使用TTS增强服务: 包含 `multiSpeakerVoiceConfig`原生Gemini格式请求会自动使用TTS增强服务:
```bash ```bash
curl -X POST "https://your-domain.com/v1beta/models/gemini-2.5-flash-preview-tts:generateContent" \ curl -X POST "https://your-domain.com/v1beta/models/gemini-2.5-flash-preview-tts:generateContent" \
@@ -107,7 +107,7 @@ curl -X POST "https://your-domain.com/v1beta/models/gemini-2.5-flash-preview-tts
-d '{ -d '{
"contents": [{ "contents": [{
"parts": [{ "parts": [{
"text": "小雅: 听众朋友们大家好!欢迎收听今天的节目。\n李想: 小雅好,听众朋友们好!今天我们来聊聊人工智能的发展。" "text": "Alice: Hello everyone, welcome to our show today.\nBob: Hi Alice, and hello to all our listeners! Today we are talking about AI development."
}] }]
}], }],
"generationConfig": { "generationConfig": {
@@ -116,18 +116,18 @@ curl -X POST "https://your-domain.com/v1beta/models/gemini-2.5-flash-preview-tts
"multiSpeakerVoiceConfig": { "multiSpeakerVoiceConfig": {
"speakerVoiceConfigs": [ "speakerVoiceConfigs": [
{ {
"speaker": "李想", "speaker": "Alice",
"voiceConfig": { "voiceConfig": {
"prebuiltVoiceConfig": { "prebuiltVoiceConfig": {
"voiceName": "Kore" "voiceName": "Puck"
} }
} }
}, },
{ {
"speaker": "小雅", "speaker": "Bob",
"voiceConfig": { "voiceConfig": {
"prebuiltVoiceConfig": { "prebuiltVoiceConfig": {
"voiceName": "Puck" "voiceName": "Kore"
} }
} }
} }
@@ -138,33 +138,28 @@ curl -X POST "https://your-domain.com/v1beta/models/gemini-2.5-flash-preview-tts
}' }'
``` ```
### 单人TTS请求(使用原有服务) ### 3. OpenAI兼容TTS请求(使用原有服务)
不包含 `multiSpeakerVoiceConfig` 的TTS请求使用原有的Gemini TTS服务 OpenAI兼容格式的TTS请求使用不同的API路径,不受本模块影响
```bash ```bash
curl -X POST "https://your-domain.com/v1beta/models/gemini-2.5-flash-preview-tts:generateContent" \ curl -X POST "https://your-domain.com/v1/audio/speech" \
-H "Content-Type: application/json" \ -H "Content-Type: application/json" \
-H "x-goog-api-key: your-token" \ -H "Authorization: Bearer your-token" \
-d '{ -d '{
"contents": [{ "model": "tts-1",
"parts": [{ "input": "这是一个OpenAI兼容格式的TTS测试。",
"text": "Hello, this is a single speaker test." "voice": "alloy"
}] }' \
}], --output openai_tts_test.wav
"generationConfig": {
"responseModalities": ["AUDIO"],
"speechConfig": {
"voiceConfig": {
"prebuiltVoiceConfig": {
"voiceName": "Kore"
}
}
}
}
}'
``` ```
**注意**OpenAI兼容TTS请求:
- 使用路径:`/v1/audio/speech`
- 使用Authorization头而不是x-goog-api-key
- 返回音频文件而不是JSON响应
- 不受本模块的TTS增强服务影响
### 普通文本生成(使用原有服务) ### 普通文本生成(使用原有服务)
非TTS模型的请求会使用原有的Gemini聊天服务,完全不受影响: 非TTS模型的请求会使用原有的Gemini聊天服务,完全不受影响: