音訊

音訊 API 涵蓋 Gemini 原生音訊理解，以及 OpenAI 相容的語音合成、音訊轉錄與音訊翻譯端點。

原生 Gemini 格式

當您需要以結構化 parts 處理多模態音訊理解或生成時，請使用 Gemini 相容的 generateContent 介面。

POST

https://api.dgrid.ai

POST/v1/models/{model}:generateContent

Authorization

Authorization: Bearer <DGRID_API_KEY>

Request

application/json

Response

200 · application/json

路徑參數

參數	型別	必填	說明
model	string	是	目標模型 ID，例如 gemini-1.5-pro。

請求體

欄位	型別	必填	說明
contents	array	是	輸入內容陣列。
contents[]	object	是	目前範例會傳送一個只包含空物件的陣列項目。
generationConfig	object	是	用於語音回應生成的設定。
generationConfig.responseModalities	array	是	要返回的模態清單。
generationConfig.speechConfig	object	是	語音生成設定。
generationConfig.speechConfig.voiceConfig	object	是	語音設定包裝物件。
generationConfig.speechConfig.voiceConfig.prebuiltVoiceConfig	object	是	預建語音設定。
generationConfig.speechConfig.voiceConfig.prebuiltVoiceConfig.voiceName	string	是	要使用的語音名稱。

回應體

欄位	型別	說明
candidates	array	模型返回的候選回應。
candidates[].content	object	生成內容物件。
candidates[].content.role	string	生成內容的角色。
candidates[].content.parts	array	返回內容片段。
candidates[].finishReason	string	範例回應返回的結束原因字串。
candidates[].safetyRatings	array	安全評分結果。
usageMetadata	object	Token 使用統計。
usageMetadata.promptTokenCount	integer	輸入 token 數。
usageMetadata.candidatesTokenCount	integer	輸出 token 數。
usageMetadata.totalTokenCount	integer	總 token 數。

請求範例

cURL / JavaScript / Go / Python / Java / C#

cURL

curl -X POST "https://api.dgrid.ai/v1/models/string:generateContent" \
  -H "Authorization: Bearer " \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [
      {}
    ],
    "generationConfig": {
      "responseModalities": [
        "string"
      ],
      "speechConfig": {
        "voiceConfig": {
          "prebuiltVoiceConfig": {
            "voiceName": "string"
          }
        }
      }
    }
  }'

JavaScript

const body = JSON.stringify({
  "contents": [
    {}
  ],
  "generationConfig": {
    "responseModalities": [
      "string"
    ],
    "speechConfig": {
      "voiceConfig": {
        "prebuiltVoiceConfig": {
          "voiceName": "string"
        }
      }
    }
  }
})

fetch("https://api.dgrid.ai/v1/models/string:generateContent", {
  method: "POST",
  headers: {
    "Content-Type": "application/json",
    "Authorization": "Bearer "
  },
  body
})

Go

package main

import (
  "fmt"
  "net/http"
  "io/ioutil"
  "strings"
)

func main() {
  url := "https://api.dgrid.ai/v1/models/string:generateContent"
  body := strings.NewReader(`{
    "contents": [
      {}
    ],
    "generationConfig": {
      "responseModalities": [
        "string"
      ],
      "speechConfig": {
        "voiceConfig": {
          "prebuiltVoiceConfig": {
            "voiceName": "string"
          }
        }
      }
    }
  }`)
  req, _ := http.NewRequest("POST", url, body)
  req.Header.Add("Authorization", "Bearer ")
  req.Header.Add("Content-Type", "application/json")
  res, _ := http.DefaultClient.Do(req)
  defer res.Body.Close()
  body, _ := ioutil.ReadAll(res.Body)

  fmt.Println(res)
  fmt.Println(string(body))
}

Python

import requests

url = "https://api.dgrid.ai/v1/models/string:generateContent"
body = """{
  "contents": [
    {}
  ],
  "generationConfig": {
    "responseModalities": [
      "string"
    ],
    "speechConfig": {
      "voiceConfig": {
        "prebuiltVoiceConfig": {
          "voiceName": "string"
        }
      }
    }
  }
}"""
response = requests.request("POST", url, data = body, headers = {
  "Content-Type": "application/json", 
  "Authorization": "Bearer "
})

print(response.text)

Java

import java.net.URI;
import java.net.http.HttpClient;
import java.net.http.HttpRequest;
import java.net.http.HttpResponse;
import java.net.http.HttpResponse.BodyHandlers;
import java.time.Duration;
import java.net.http.HttpRequest.BodyPublishers;

var body = BodyPublishers.ofString("""{
  "contents": [
    {}
  ],
  "generationConfig": {
    "responseModalities": [
      "string"
    ],
    "speechConfig": {
      "voiceConfig": {
        "prebuiltVoiceConfig": {
          "voiceName": "string"
        }
      }
    }
  }
}""");
HttpClient client = HttpClient.newBuilder()
  .connectTimeout(Duration.ofSeconds(10))
  .build();

HttpRequest.Builder requestBuilder = HttpRequest.newBuilder()
  .uri(URI.create("https://api.dgrid.ai/v1/models/string:generateContent"))
  .header("Authorization", "Bearer ")
  .header("Content-Type", "application/json")
  .POST(body)
  .build();

try {
  HttpResponse<String> response = client.send(requestBuilder.build(), BodyHandlers.ofString());
  System.out.println("Status code: " + response.statusCode());
  System.out.println("Response body: " + response.body());
} catch (Exception e) {
  e.printStackTrace();
}

C#

using System;
using System.Net.Http;
using System.Text;

var body = new StringContent("""
{
  "contents": [
    {}
  ],
  "generationConfig": {
    "responseModalities": [
      "string"
    ],
    "speechConfig": {
      "voiceConfig": {
        "prebuiltVoiceConfig": {
          "voiceName": "string"
        }
      }
    }
  }
}
""", Encoding.UTF8, "application/json");

var client = new HttpClient();
client.DefaultRequestHeaders.Add("Authorization", "Bearer ");
var response = await client.PostAsync("https://api.dgrid.ai/v1/models/string:generateContent", body);
var responseBody = await response.Content.ReadAsStringAsync();

200 回應

application/json

{
  "candidates": [
    {
      "content": {
        "role": "string",
        "parts": [
          {}
        ]
      },
      "finishReason": "string",
      "safetyRatings": [
        {}
      ]
    }
  ],
  "usageMetadata": {
    "promptTokenCount": 0,
    "candidatesTokenCount": 0,
    "totalTokenCount": 0
  }
}

文字轉語音

透過 OpenAI 相容 speech 介面，將文字轉換為自然語音。

POST

https://api.dgrid.ai

POST/v1/audio/speech

Authorization

Authorization: Bearer <DGRID_API_KEY>

Request

application/json

Response

200 · audio/mpeg

請求體

欄位	型別	必填	預設值	說明
model	string	是	-	語音模型，例如 tts-1 或 tts-1-hd。
input	string	是	-	要合成的文字，最長 4096 字元。
voice	string	是	-	語音預設，例如 alloy、echo、fable、onyx、nova、shimmer。
response_format	string	否	mp3	輸出音訊格式。
speed	number	否	1.0	語速，範圍 0.25 到 4.0。

回應

此端點返回二進位音訊串流，請直接將回應內容保存為檔案或上傳到儲存目標。

請求範例

cURL / JavaScript / Go / Python / Java / C#

cURL

curl -X POST "https://api.dgrid.ai/v1/audio/speech" \
  -H "Authorization: Bearer " \
  -H "Content-Type: application/json" \
  -d '{
    "model": "tts-1",
    "input": "string",
    "voice": "alloy"
  }'

JavaScript

const body = JSON.stringify({
  "model": "tts-1",
  "input": "string",
  "voice": "alloy"
})

fetch("https://api.dgrid.ai/v1/audio/speech", {
  method: "POST",
  headers: {
    "Content-Type": "application/json",
    "Authorization": "Bearer "
  },
  body
})

Go

package main

import (
  "fmt"
  "net/http"
  "io/ioutil"
  "strings"
)

func main() {
  url := "https://api.dgrid.ai/v1/audio/speech"
  body := strings.NewReader(`{
    "model": "tts-1",
    "input": "string",
    "voice": "alloy"
  }`)
  req, _ := http.NewRequest("POST", url, body)
  req.Header.Add("Authorization", "Bearer ")
  req.Header.Add("Content-Type", "application/json")
  res, _ := http.DefaultClient.Do(req)
  defer res.Body.Close()
  body, _ := ioutil.ReadAll(res.Body)

  fmt.Println(res)
  fmt.Println(string(body))
}

Python

import requests

url = "https://api.dgrid.ai/v1/audio/speech"
body = """{
  "model": "tts-1",
  "input": "string",
  "voice": "alloy"
}"""
response = requests.request("POST", url, data = body, headers = {
  "Content-Type": "application/json", 
  "Authorization": "Bearer "
})

print(response.text)

Java

import java.net.URI;
import java.net.http.HttpClient;
import java.net.http.HttpRequest;
import java.net.http.HttpResponse;
import java.net.http.HttpResponse.BodyHandlers;
import java.time.Duration;
import java.net.http.HttpRequest.BodyPublishers;

var body = BodyPublishers.ofString("""{
  "model": "tts-1",
  "input": "string",
  "voice": "alloy"
}""");
HttpClient client = HttpClient.newBuilder()
  .connectTimeout(Duration.ofSeconds(10))
  .build();

HttpRequest.Builder requestBuilder = HttpRequest.newBuilder()
  .uri(URI.create("https://api.dgrid.ai/v1/audio/speech"))
  .header("Authorization", "Bearer ")
  .header("Content-Type", "application/json")
  .POST(body)
  .build();

try {
  HttpResponse<String> response = client.send(requestBuilder.build(), BodyHandlers.ofString());
  System.out.println("Status code: " + response.statusCode());
  System.out.println("Response body: " + response.body());
} catch (Exception e) {
  e.printStackTrace();
}

C#

using System;
using System.Net.Http;
using System.Text;

var body = new StringContent("""
{
  "model": "tts-1",
  "input": "string",
  "voice": "alloy"
}
""", Encoding.UTF8, "application/json");

var client = new HttpClient();
client.DefaultRequestHeaders.Add("Authorization", "Bearer ");
var response = await client.PostAsync("https://api.dgrid.ai/v1/audio/speech", body);
var responseBody = await response.Content.ReadAsStringAsync();

回應

binary stream

"string"

音訊轉錄

透過 OpenAI 相容 Whisper 風格介面，將上傳音訊轉為文字。

POST

https://api.dgrid.ai

POST/v1/audio/transcriptions

Authorization

Authorization: Bearer <DGRID_API_KEY>

Request

multipart/form-data

Response

200 · application/json

表單欄位

欄位	型別	必填	預設值	說明
file	file	是	-	音訊檔，大小上限 25 MB。
model	string	是	-	模型 ID，例如 whisper-1。
language	string	否	-	ISO-639-1 語言代碼，例如 zh、en、ko。
prompt	string	否	-	用於偏置轉錄結果的提示文字。
response_format	string	否	json	json、text、srt、verbose_json 或 vtt。
temperature	number	否	0	採樣溫度，範圍 0 到 1。

支援格式

FLAC、MP3、MP4、MPEG、MPGA、M4A、OGG、WAV 與 WebM。

回應體

欄位	型別	說明
text	string	轉錄後的文字內容。

若 response_format 設為 verbose_json，回應中還會包含 task、language、duration 與逐段時間軸資訊。

請求範例

cURL / JavaScript / Go / Python / Java / C#

cURL

curl -X POST "https://api.dgrid.ai/v1/audio/transcriptions" \
  -H "Authorization: Bearer " \
  -F file="string" \
  -F model="whisper-1"

JavaScript

const body = new FormData();
body.set(file, "string")
body.set(model, "whisper-1")

fetch("https://api.dgrid.ai/v1/audio/transcriptions", {
  method: "POST",
  headers: {
    "Content-Type": "multipart/form-data",
    "Authorization": "Bearer "
  },
  body
})

Go

package main

import (
  "fmt"
  "net/http"
  "io/ioutil"
  "mime/multipart"
  "bytes"
)

func main() {
  url := "https://api.dgrid.ai/v1/audio/transcriptions"
  body := new(bytes.Buffer)
  mp := multipart.NewWriter(payload)
  mp.WriteField("file", `string`)
  mp.WriteField("model", `whisper-1`)
  req, _ := http.NewRequest("POST", url, body)
  req.Header.Add("Authorization", "Bearer ")
  req.Header.Add("Content-Type", "multipart/form-data")
  res, _ := http.DefaultClient.Do(req)
  defer res.Body.Close()
  body, _ := ioutil.ReadAll(res.Body)

  fmt.Println(res)
  fmt.Println(string(body))
}

Python

import requests

url = "https://api.dgrid.ai/v1/audio/transcriptions"
body = {
  "file": "string",
  "model": "whisper-1"
}
response = requests.request("POST", url, data = body, headers = {
  "Content-Type": "multipart/form-data", 
  "Authorization": "Bearer "
})

print(response.text)

Java

import java.net.URI;
import java.net.http.HttpClient;
import java.net.http.HttpRequest;
import java.net.http.HttpResponse;
import java.net.http.HttpResponse.BodyHandlers;
import java.time.Duration;
import java.net.http.HttpRequest.BodyPublishers;

var body = BodyPublishers.ofByteArray(new byte[] { ... });
HttpClient client = HttpClient.newBuilder()
  .connectTimeout(Duration.ofSeconds(10))
  .build();

HttpRequest.Builder requestBuilder = HttpRequest.newBuilder()
  .uri(URI.create("https://api.dgrid.ai/v1/audio/transcriptions"))
  .header("Authorization", "Bearer ")
  .header("Content-Type", "multipart/form-data")
  .POST(body)
  .build();

try {
  HttpResponse<String> response = client.send(requestBuilder.build(), BodyHandlers.ofString());
  System.out.println("Status code: " + response.statusCode());
  System.out.println("Response body: " + response.body());
} catch (Exception e) {
  e.printStackTrace();
}

C#

using System;
using System.Net.Http;
using System.Text;

var body = new MultipartFormDataContent();

var client = new HttpClient();
client.DefaultRequestHeaders.Add("Authorization", "Bearer ");
var response = await client.PostAsync("https://api.dgrid.ai/v1/audio/transcriptions", body);
var responseBody = await response.Content.ReadAsStringAsync();

200 回應

application/json

{
  "text": "string"
}

音訊翻譯

透過 OpenAI 相容翻譯端點，將上傳音訊翻譯為英文。

POST

https://api.dgrid.ai

POST/v1/audio/translations

Authorization

Authorization: Bearer <DGRID_API_KEY>

Request

multipart/form-data

Response

200 · application/json

表單欄位

欄位	型別	必填	預設值	說明
file	file	是	-	原始音訊檔。
model	string	是	-	模型 ID，例如 whisper-1。
prompt	string	否	-	可選英文提示文字。
response_format	string	否	json	json、text、srt、verbose_json 或 vtt。
temperature	number	否	0	採樣溫度，範圍 0 到 1。

回應體

欄位	型別	說明
text	string	翻譯後的英文文字。

請求範例

cURL / JavaScript / Go / Python / Java / C#

cURL

curl -X POST "https://api.dgrid.ai/v1/audio/translations" \
  -H "Authorization: Bearer " \
  -F file="string" \
  -F model="string"

JavaScript

const body = new FormData();
body.set(file, "string")
body.set(model, "string")

fetch("https://api.dgrid.ai/v1/audio/translations", {
  method: "POST",
  headers: {
    "Content-Type": "multipart/form-data",
    "Authorization": "Bearer "
  },
  body
})

Go

package main

import (
  "fmt"
  "net/http"
  "io/ioutil"
  "mime/multipart"
  "bytes"
)

func main() {
  url := "https://api.dgrid.ai/v1/audio/translations"
  body := new(bytes.Buffer)
  mp := multipart.NewWriter(payload)
  mp.WriteField("file", `string`)
  mp.WriteField("model", `string`)
  req, _ := http.NewRequest("POST", url, body)
  req.Header.Add("Authorization", "Bearer ")
  req.Header.Add("Content-Type", "multipart/form-data")
  res, _ := http.DefaultClient.Do(req)
  defer res.Body.Close()
  body, _ := ioutil.ReadAll(res.Body)

  fmt.Println(res)
  fmt.Println(string(body))
}

Python

import requests

url = "https://api.dgrid.ai/v1/audio/translations"
body = {
  "file": "string",
  "model": "string"
}
response = requests.request("POST", url, data = body, headers = {
  "Content-Type": "multipart/form-data", 
  "Authorization": "Bearer "
})

print(response.text)

Java

import java.net.URI;
import java.net.http.HttpClient;
import java.net.http.HttpRequest;
import java.net.http.HttpResponse;
import java.net.http.HttpResponse.BodyHandlers;
import java.time.Duration;
import java.net.http.HttpRequest.BodyPublishers;

var body = BodyPublishers.ofByteArray(new byte[] { ... });
HttpClient client = HttpClient.newBuilder()
  .connectTimeout(Duration.ofSeconds(10))
  .build();

HttpRequest.Builder requestBuilder = HttpRequest.newBuilder()
  .uri(URI.create("https://api.dgrid.ai/v1/audio/translations"))
  .header("Authorization", "Bearer ")
  .header("Content-Type", "multipart/form-data")
  .POST(body)
  .build();

try {
  HttpResponse<String> response = client.send(requestBuilder.build(), BodyHandlers.ofString());
  System.out.println("Status code: " + response.statusCode());
  System.out.println("Response body: " + response.body());
} catch (Exception e) {
  e.printStackTrace();
}

C#

using System;
using System.Net.Http;
using System.Text;

var body = new MultipartFormDataContent();

var client = new HttpClient();
client.DefaultRequestHeaders.Add("Authorization", "Bearer ");
var response = await client.PostAsync("https://api.dgrid.ai/v1/audio/translations", body);
var responseBody = await response.Content.ReadAsStringAsync();

200 回應

application/json

{
  "text": "string"
}