Skip to content

tts-guide.md

Latest commit

 

History

History
270 lines (226 loc) · 6.66 KB

tts-guide.md

File metadata and controls

270 lines (226 loc) · 6.66 KB

TTS 语音生成使用指南

本指南介绍如何使用 AIStudio2API 的 TTS (Text-to-Speech) 功能生成语音。

支持的模型

  • gemini-2.5-flash-preview-tts
  • gemini-2.5-pro-preview-tts

端点

  • POST /generate-speech
  • POST /v1beta/models/{model}:generateContent (兼容官方 API)

单说话人语音生成

基础示例

curl -X POST http://localhost:2048/generate-speech \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gemini-2.5-flash-preview-tts",
    "contents": "Hello, this is a test.",
    "generationConfig": {
      "responseModalities": ["AUDIO"],
      "speechConfig": {
        "voiceConfig": {
          "prebuiltVoiceConfig": {"voiceName": "Kore"}
        }
      }
    }
  }'

Python 示例

import requests
import base64

url = "http://localhost:2048/generate-speech"
payload = {
    "model": "gemini-2.5-flash-preview-tts",
    "contents": "Hello, this is a test.",
    "generationConfig": {
        "responseModalities": ["AUDIO"],
        "speechConfig": {
            "voiceConfig": {
                "prebuiltVoiceConfig": {"voiceName": "Kore"}
            }
        }
    }
}

response = requests.post(url, json=payload)
data = response.json()

# 提取 Base64 音频数据
audio_base64 = data['candidates'][0]['content']['parts'][0]['inlineData']['data']

# 保存为 WAV 文件
with open('output.wav', 'wb') as f:
    f.write(base64.b64decode(audio_base64))

PowerShell 示例

$url = "http://localhost:2048/generate-speech"
$body = @{
    model = "gemini-2.5-flash-preview-tts"
    contents = "Hello world"
    generationConfig = @{
        responseModalities = @("AUDIO")
        speechConfig = @{
            voiceConfig = @{
                prebuiltVoiceConfig = @{
                    voiceName = "Kore"
                }
            }
        }
    }
} | ConvertTo-Json -Depth 5

$response = Invoke-RestMethod -Uri $url -Method Post -ContentType "application/json" -Body $body
$base64Audio = $response.candidates[0].content.parts[0].inlineData.data
[System.IO.File]::WriteAllBytes("C:\output.wav", [Convert]::FromBase64String($base64Audio))

多说话人语音生成

基础示例

curl -X POST http://localhost:2048/generate-speech \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gemini-2.5-flash-preview-tts",
    "contents": "Joe: How are you today?\nJane: I am doing great, thanks!",
    "generationConfig": {
      "responseModalities": ["AUDIO"],
      "speechConfig": {
        "multiSpeakerVoiceConfig": {
          "speakerVoiceConfigs": [
            {
              "speaker": "Joe",
              "voiceConfig": {
                "prebuiltVoiceConfig": {"voiceName": "Kore"}
              }
            },
            {
              "speaker": "Jane",
              "voiceConfig": {
                "prebuiltVoiceConfig": {"voiceName": "Puck"}
              }
            }
          ]
        }
      }
    }
  }'

Python 示例

import requests
import base64

url = "http://localhost:2048/generate-speech"
payload = {
    "model": "gemini-2.5-flash-preview-tts",
    "contents": "Joe: How are you?\nJane: I'm fine, thanks!",
    "generationConfig": {
        "responseModalities": ["AUDIO"],
        "speechConfig": {
            "multiSpeakerVoiceConfig": {
                "speakerVoiceConfigs": [
                    {
                        "speaker": "Joe",
                        "voiceConfig": {
                            "prebuiltVoiceConfig": {"voiceName": "Kore"}
                        }
                    },
                    {
                        "speaker": "Jane",
                        "voiceConfig": {
                            "prebuiltVoiceConfig": {"voiceName": "Puck"}
                        }
                    }
                ]
            }
        }
    }
}

response = requests.post(url, json=payload)
data = response.json()
audio_base64 = data['candidates'][0]['content']['parts'][0]['inlineData']['data']

with open('conversation.wav', 'wb') as f:
    f.write(base64.b64decode(audio_base64))

可用语音列表

共 30 种预置语音可选：

语音名称 风格 语音名称 风格
Zephyr Bright Puck Upbeat
Charon Informative Kore Firm
Fenrir Excitable Leda Youthful
Orus Firm Aoede Breezy
Callirrhoe Easy-going Autonoe Bright
Enceladus Breathy Iapetus Clear
Umbriel Easy-going Algieba Smooth
Despina Smooth Erinome Clear
Algenib Gravelly Rasalgethi Informative
Laomedeia Upbeat Achernar Soft
Alnilam Firm Schedar Even
Gacrux Mature Pulcherrima Forward
Achird Friendly Zubenelgenubi Casual
Vindemiatrix Gentle Sadachbia Lively
Sadaltager Knowledgeable Sulafat Warm

响应格式

成功的响应格式如下：

{
  "candidates": [{
    "content": {
      "parts": [{
        "inlineData": {
          "mimeType": "audio/wav",
          "data": "<Base64 编码的 WAV 音频数据>"
        }
      }],
      "role": "model"
    },
    "finishReason": "STOP",
    "index": 0
  }],
  "usageMetadata": {
    "promptTokenCount": 3,
    "candidatesTokenCount": 0,
    "totalTokenCount": 3
  },
  "modelVersion": "gemini-2.5-flash-preview-tts"
}

支持的语言

TTS 模型自动检测输入语言，支持以下 24 种语言：

  • 阿拉伯语（埃及）ar-EG
  • 英语（美国）en-US
  • 法语（法国）fr-FR
  • 德语（德国）de-DE
  • 西班牙语（美国）es-US
  • 印地语（印度）hi-IN
  • 印尼语（印度尼西亚）id-ID
  • 意大利语（意大利）it-IT
  • 日语（日本）ja-JP
  • 韩语（韩国）ko-KR
  • 葡萄牙语（巴西）pt-BR
  • 俄语（俄罗斯）ru-RU
  • 荷兰语（荷兰）nl-NL
  • 波兰语（波兰）pl-PL
  • 泰语（泰国）th-TH
  • 土耳其语（土耳其）tr-TR
  • 越南语（越南）vi-VN
  • 罗马尼亚语（罗马尼亚）ro-RO
  • 乌克兰语（乌克兰）uk-UA
  • 孟加拉语（孟加拉国）bn-BD
  • 英语（印度）en-IN
  • 马拉地语（印度）mr-IN
  • 泰米尔语（印度）ta-IN
  • 泰卢固语（印度）te-IN

限制

  • TTS 模型仅接受文本输入，生成音频输出
  • 上下文窗口限制为 32k tokens
  • 音频格式为 WAV，采样率 24kHz，单声道，16-bit PCM

故障排除

音频无法播放

确保正确解码 Base64 数据：

import base64
audio_data = base64.b64decode(base64_string)

语音名称无效

检查语音名称拼写是否正确，区分大小写。

模型不可用

确保使用的是 TTS 专用模型：

  • gemini-2.5-flash-preview-tts
  • gemini-2.5-pro-preview-tts