Desain suara

Desain suara memungkinkan Anda membuat persona vokal persisten yang benar-benar baru dari deskripsi bahasa alami menggunakan endpoint Suara Gemini API (POST /v1beta/voices). Daripada terbatas pada suara bawaan atau merekam audio referensi, Anda dapat mendeskripsikan usia, timbre vokal, aksen, dan penyampaian dasar karakter, serta menerima ID voice_... yang dapat digunakan kembali dan disimpan ke project Anda.

Cara tercepat untuk mendesain, menguji, dan melakukan iterasi pada suara kustom adalah dengan studio Desain Suara interaktif di Google AI Studio. Anda dapat membuat persona kustom dari perintah teks, mengujinya dengan skrip contoh, dan menyalin ID voice_... yang dihasilkan langsung ke kode aplikasi Anda.

Gemini 3.8 Flash TTS (gemini-3.8-flash-tts) dan Gemini 3.8 Flash-Lite TTS (gemini-3.8-flash-lite-tts) mendukung Desain suara.

Membuat suara yang didesain

Gunakan Google GenAI SDK (google-genai 2.25.0+ / @google/genai 2.24.0+) atau REST API untuk membuat suara kustom dari deskripsi teks. Untuk suara "prompted", voices.create (CreateVoice) dan voices.get (GetVoice) menampilkan kolom sample_audio hanya output (mime_type: "audio/wav", data yang dienkode base64) sehingga Anda dapat langsung mencoba suara yang dihasilkan:

Python

import base64
from google import genai

client = genai.Client()

# 1. Design a custom voice persona from natural language
created_voice = client.voices.create(
    store=True,
    voice={
        "model": "gemini-3.8-flash-tts",
        "type": "prompted",
        "display_name": "Warm British Astronomer",
        "gender": "male",
        "language_code": "en-GB",
        "prompted": {
            "input": (
                "A warm, thoughtful astronomer in his late 60s with a gentle"
                " British accent, speaking with quiet wonder."
            )
        },
    },
)

print(f"Created voice ID: {created_voice.id}")

# Save the generated sample_audio preview (audio/wav) returned by CreateVoice
if created_voice.sample_audio and created_voice.sample_audio.data:
    with open("voice_preview.wav", "wb") as f:
        f.write(base64.b64decode(created_voice.sample_audio.data))

JavaScript

import * as fs from "node:fs";
import { GoogleGenAI } from "@google/genai";

const ai = new GoogleGenAI();

// 1. Design a custom voice persona from natural language
const createdVoice = await ai.voices.create({
  store: true,
  voice: {
    model: "gemini-3.8-flash-tts",
    type: "prompted",
    display_name: "Warm British Astronomer",
    gender: "male",
    language_code: "en-GB",
    prompted: {
      input:
        "A warm, thoughtful astronomer in his late 60s with a gentle British accent, speaking with quiet wonder.",
    },
  },
});

console.log(`Created voice ID: ${createdVoice.id}`);

// Save the generated sample_audio preview (audio/wav) returned by CreateVoice
if (createdVoice.sample_audio?.data) {
  fs.writeFileSync(
    "voice_preview.wav",
    Buffer.from(createdVoice.sample_audio.data, "base64")
  );
}

REST

curl "https://generativelanguage.googleapis.com/v1beta/voices" \
  -H "x-goog-api-key: $GEMINI_API_KEY" \
  -H "Content-Type: application/json" \
  -X POST \
  -d '{
    "store": true,
    "voice": {
      "model": "gemini-3.8-flash-tts",
      "type": "prompted",
      "display_name": "Warm British Astronomer",
      "gender": "male",
      "language_code": "en-GB",
      "prompted": {
        "input": "A warm, thoughtful astronomer in his late 60s with a gentle British accent, speaking with quiet wonder."
      }
    }
  }' | tee created_voice.json | jq -r '.sample_audio.data' | base64 --decode > voice_preview.wav

Cara kerja desain Voice

  1. Membuat suara yang dipicu: Panggil voices.create (POST /v1beta/voices) dengan type="prompted" dan store=True.
  2. Menerima pratinjau voice_id dan sample_audio persisten: API membuat identitas vokal, menyimpannya di project Anda, dan menampilkan ID permanen (misalnya, voice_abc123...) bersama dengan sample_audio (mime_type: "audio/wav", data berenkode base64) yang berisi audio pratinjau yang dihasilkan untuk suara.
  3. Sintesis ucapan: Teruskan voice_id di mana pun nama suara diterima dalam permintaan sintesis Anda.

Menyintesis ucapan dengan suara yang Anda rancang

Setelah membuat suara, teruskan id (voice_...) ke Interactions API untuk menghasilkan ucapan:

Python

import base64
from google import genai

client = genai.Client()

interaction = client.interactions.create(
    model="gemini-3.8-flash-tts",
    input=[{
        "type": "user_input",
        "content": [{
            "type": "text",
            "text": (
                "Look out past the rings of Saturn. Those faint photons left"
                " their source millions of years ago."
            ),
            "annotations": [{
                "type": "speech_metadata",
                "style": "reflective and awe-inspired",
            }],
        }],
    }],
    response_format={"type": "audio"},
    generation_config={
        "speech_config": [
            {"voice": created_voice.id},
        ]
    },
)

with open("designed_voice.wav", "wb") as f:
    f.write(base64.b64decode(interaction.output_audio.data))

JavaScript

import * as fs from "node:fs";
import { GoogleGenAI } from "@google/genai";

const ai = new GoogleGenAI();

const interaction = await ai.interactions.create({
  model: "gemini-3.8-flash-tts",
  input: [{
    type: "user_input",
    content: [{
      type: "text",
      text: "Look out past the rings of Saturn. Those faint photons left their source millions of years ago.",
      annotations: [{
        type: "speech_metadata",
        style: "reflective and awe-inspired",
      }],
    }],
  }],
  response_format: { type: "audio" },
  generation_config: {
    speech_config: [
      { voice: createdVoice.id },
    ],
  },
});

fs.writeFileSync("designed_voice.wav", Buffer.from(interaction.output_audio.data, "base64"));

REST

curl "https://generativelanguage.googleapis.com/v1beta/interactions" \
  -H "x-goog-api-key: $GEMINI_API_KEY" \
  -H "Content-Type: application/json" \
  -X POST \
  -d '{
    "model": "gemini-3.8-flash-tts",
    "input": [{
      "type": "user_input",
      "content": [{
        "type": "text",
        "text": "Look out past the rings of Saturn. Those faint photons left their source millions of years ago.",
        "annotations": [{
          "type": "speech_metadata",
          "style": "reflective and awe-inspired"
        }]
      }]
    }],
    "response_format": {"type": "audio"},
    "generation_config": {
      "speech_config": [
        {"voice": "voice_YOUR_DESIGNED_VOICE_ID"}
      ]
    }
  }' | jq -r '[.steps[] | select(.type=="model_output") | .content[] | select(.type=="audio")] | last | .data' | base64 --decode > out.wav

Mengelola suara Anda

Anda dapat mencantumkan, memfilter, memeriksa, dan menghapus suara tersimpan Anda kapan saja menggunakan Voices API (lihat Library Suara yang Diperluas dan pemfilteran untuk semua parameter filter).

  • Batas penyimpanan dan TTL: Suara stateful (store=True, dibagikan di seluruh suara yang diminta dan direplikasi) memiliki batas 200 suara per project dan TTL 1 tahun (time-to-live).

  • Ketersediaan sample_audio: voices.create() (CreateVoice) dan voices.get() (GetVoice) mengisi sample_audio (mime_type: "audio/wav", data berenkode base64) untuk suara "prompted". Agar listingan tetap ringan, voices.list() (ListVoices) menghilangkan sample_audio (dan sample_audio tidak disetel untuk suara "replicated" dan "prebuilt").

Python

from google import genai

client = genai.Client()

# List stored prompted voices in your project filtered by language
response = client.voices.list(
    type_=["prompted"],
    language_code=["en-US", "en-GB"],
)
for voice in response.voices or []:
    print(voice.id, voice.display_name, voice.type)

# Retrieve a specific voice by ID
voice_details = client.voices.get(id=created_voice.id)

# Delete a stored custom voice
client.voices.delete(id=created_voice.id)

JavaScript

import { GoogleGenAI } from "@google/genai";

const ai = new GoogleGenAI();

// List stored prompted voices in your project filtered by language
const response = await ai.voices.list({
  type: ["prompted"],
  language_code: ["en-US", "en-GB"],
});
for (const voice of response.voices ?? []) {
  console.log(voice.id, voice.display_name, voice.type);
}

// Retrieve a specific voice by ID
const voiceDetails = await ai.voices.get(createdVoice.id);

// Delete a stored custom voice
await ai.voices.delete(createdVoice.id);

REST

# List stored prompted voices filtered by language
curl -G "https://generativelanguage.googleapis.com/v1beta/voices" \
  -H "x-goog-api-key: $GEMINI_API_KEY" \
  --data-urlencode "type=prompted" \
  --data-urlencode "language_code=en-US" \
  --data-urlencode "language_code=en-GB"

# Retrieve a specific voice by ID
curl "https://generativelanguage.googleapis.com/v1beta/voices/voice_YOUR_DESIGNED_VOICE_ID" \
  -H "x-goog-api-key: $GEMINI_API_KEY"

# Delete a stored custom voice
curl -X DELETE "https://generativelanguage.googleapis.com/v1beta/voices/voice_YOUR_DESIGNED_VOICE_ID" \
  -H "x-goog-api-key: $GEMINI_API_KEY"

Praktik terbaik perintah untuk desain Voice

  • Masukkan ciri vokal permanen dalam Desain suara, bukan style: Tentukan karakteristik tetap—seperti usia, gender, timbre, karakter suara, dan aksen regional—saat membuat suara di voices.create.
  • Cadangkan speech_metadata.style untuk emosi situasional: Setelah suara kustom Anda dibuat, gunakan perintah style singkat (misalnya, "whispered urgently" atau "cheerful and energetic") untuk mengarahkan akting belokan demi belokan tanpa mengubah identitas inti pembicara.
  • Buat perintah yang spesifik dan ringkas: Deskripsi 1–2 kalimat yang jelas (seperti "Seorang komentator olahraga yang bersemangat dan fasih di usia 30-an dengan sedikit aksen Midwest") akan menghasilkan hasil yang lebih bersih dan konsisten daripada paragraf yang bertentangan atau terlalu panjang.

Langkah berikutnya