Gerar músicas com o Lyria 3

O Lyria 3 é a família de modelos de geração de música do Google, disponível na API Gemini. Com o Lyria 3, é possível gerar áudio estéreo de alta qualidade (48 kHz) com comandos de texto ou imagens. Esses modelos oferecem coerência estrutural, incluindo vocais, letras sincronizadas e arranjos instrumentais completos.

A família Lyria 3 inclui dois modelos:

Modelo ID do modelo Ideal para Duração Saída
Lyria 3 Clip lyria-3-clip-preview Clipes curtos, loops, prévias 30 segundos MP3
Lyria 3 Pro lyria-3-pro-preview Músicas completas com versos, refrões, pontes Alguns minutos (controláveis por comando) MP3, WAV

Os dois modelos podem ser usados com o método generateContent padrão e a nova API Interactions, que aceita entradas multimodais (texto e imagens) e produz áudio estéreo de alta fidelidade de 48 kHz.

Gerar um clipe de música

O modelo Lyria 3 Clip sempre gera um clipe de 30 segundos. Para gerar um clipe, chame o generateContent método e defina response_modalities como ["AUDIO", "TEXT"]. A inclusão de TEXT permite receber a letra ou a estrutura da música gerada junto com o áudio.

Python

from google import genai
from google.genai import types

client = genai.Client()

response = client.models.generate_content(
    model="lyria-3-clip-preview",
    contents="Create a 30-second cheerful acoustic folk song with "
             "guitar and harmonica.",
    config=types.GenerateContentConfig(
        response_modalities=["AUDIO", "TEXT"],
    ),
)

# Parse the response
for part in response.parts:
    if part.text is not None:
        print(part.text)
    elif part.inline_data is not None:
        with open("clip.mp3", "wb") as f:
            f.write(part.inline_data.data)
        print("Audio saved to clip.mp3")

JavaScript

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

const ai = new GoogleGenAI({});

async function main() {
  const response = await ai.models.generateContent({
    model: "lyria-3-clip-preview",
    contents: "Create a 30-second cheerful acoustic folk song with " +
              "guitar and harmonica.",
    config: {
      responseModalities: ["AUDIO", "TEXT"],
    },
  });

  for (const part of response.candidates[0].content.parts) {
    if (part.text) {
      console.log(part.text);
    } else if (part.inlineData) {
      const buffer = Buffer.from(part.inlineData.data, "base64");
      fs.writeFileSync("clip.mp3", buffer);
      console.log("Audio saved to clip.mp3");
    }
  }
}

main();

Go

package main

import (
    "context"
    "fmt"
    "log"
    "os"

    "google.golang.org/genai"
)

func main() {
    ctx := context.Background()
    client, err := genai.NewClient(ctx, nil)
    if err != nil {
        log.Fatal(err)
    }

    config := &genai.GenerateContentConfig{
        ResponseModalities: []string{"AUDIO", "TEXT"},
    }

    result, err := client.Models.GenerateContent(
        ctx,
        "lyria-3-clip-preview",
        genai.Text("Create a 30-second cheerful acoustic folk song " +
                   "with guitar and harmonica."),
        config,
    )
    if err != nil {
        log.Fatal(err)
    }

    for _, part := range result.Candidates[0].Content.Parts {
        if part.Text != "" {
            fmt.Println(part.Text)
        } else if part.InlineData != nil {
            err := os.WriteFile("clip.mp3", part.InlineData.Data, 0644)
            if err != nil {
                log.Fatal(err)
            }
            fmt.Println("Audio saved to clip.mp3")
        }
    }
}

Java

import com.google.genai.Client;
import com.google.genai.types.GenerateContentConfig;
import com.google.genai.types.GenerateContentResponse;
import com.google.genai.types.Part;

import java.io.IOException;
import java.nio.file.Files;
import java.nio.file.Paths;

public class GenerateMusicClip {
  public static void main(String[] args) throws IOException {

    try (Client client = new Client()) {
      GenerateContentConfig config = GenerateContentConfig.builder()
          .responseModalities("AUDIO", "TEXT")
          .build();

      GenerateContentResponse response = client.models.generateContent(
          "lyria-3-clip-preview",
          "Create a 30-second cheerful acoustic folk song with "
              + "guitar and harmonica.",
          config);

      for (Part part : response.parts()) {
        if (part.text().isPresent()) {
          System.out.println(part.text().get());
        } else if (part.inlineData().isPresent()) {
          var blob = part.inlineData().get();
          if (blob.data().isPresent()) {
            Files.write(Paths.get("clip.mp3"), blob.data().get());
            System.out.println("Audio saved to clip.mp3");
          }
        }
      }
    }
  }
}

REST

curl -s -X POST \
  "https://generativelanguage.googleapis.com/v1beta/models/lyria-3-clip-preview:generateContent" \
  -H "x-goog-api-key: $GEMINI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [{
      "parts": [
        {"text": "Create a 30-second cheerful acoustic folk song with guitar and harmonica."}
      ]
    }],
    "generationConfig": {
      "responseModalities": ["AUDIO", "TEXT"]
    }
  }'

C#

using System.Threading.Tasks;
using Google.GenAI;
using Google.GenAI.Types;
using System.IO;

public class GenerateMusicClip {
  public static async Task main() {
    var client = new Client();
    var config = new GenerateContentConfig {
      ResponseModalities = { "AUDIO", "TEXT" }
    };

    var response = await client.Models.GenerateContentAsync(
      model: "lyria-3-clip-preview",
      contents: "Create a 30-second cheerful acoustic folk song with guitar and harmonica.",
      config: config
    );

    foreach (var part in response.Candidates[0].Content.Parts) {
      if (part.Text != null) {
        Console.WriteLine(part.Text);
      } else if (part.InlineData != null) {
        await File.WriteAllBytesAsync("clip.mp3", part.InlineData.Data);
        Console.WriteLine("Audio saved to clip.mp3");
      }
    }
  }
}

Gerar uma música completa

Use o modelo lyria-3-pro-preview para gerar músicas completas que duram alguns minutos. O modelo Pro entende a estrutura musical e pode criar composições com versos, refrões e pontes distintos. É possível influenciar a duração especificando-a no comando (por exemplo, "criar uma música de 2 minutos") ou usando marcações de tempo para definir a estrutura.

Python

response = client.models.generate_content(
    model="lyria-3-pro-preview",
    contents="An epic cinematic orchestral piece about a journey home. "
             "Starts with a solo piano intro, builds through sweeping "
             "strings, and climaxes with a massive wall of sound.",
    config=types.GenerateContentConfig(
        response_modalities=["AUDIO", "TEXT"],
    ),
)

JavaScript

const response = await ai.models.generateContent({
  model: "lyria-3-pro-preview",
  contents: "An epic cinematic orchestral piece about a journey home. " +
            "Starts with a solo piano intro, builds through sweeping " +
            "strings, and climaxes with a massive wall of sound.",
  config: {
    responseModalities: ["AUDIO", "TEXT"],
  },
});

Go

result, err := client.Models.GenerateContent(
    ctx,
    "lyria-3-pro-preview",
    genai.Text("An epic cinematic orchestral piece about a journey " +
               "home. Starts with a solo piano intro, builds through " +
               "sweeping strings, and climaxes with a massive wall of sound."),
    config,
)

Java

GenerateContentResponse response = client.models.generateContent(
    "lyria-3-pro-preview",
    "An epic cinematic orchestral piece about a journey home. "
        + "Starts with a solo piano intro, builds through sweeping "
        + "strings, and climaxes with a massive wall of sound.",
    config);

REST

curl -s -X POST \
  "https://generativelanguage.googleapis.com/v1beta/models/lyria-3-pro-preview:generateContent" \
  -H "x-goog-api-key: $GEMINI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [{
      "parts": [
        {"text": "An epic cinematic orchestral piece about a journey home. Starts with a solo piano intro, builds through sweeping strings, and climaxes with a massive wall of sound."}
      ]
    }],
    "generationConfig": {
      "responseModalities": ["AUDIO", "TEXT"]
    }
  }'

C#

var response = await client.Models.GenerateContentAsync(
  model: "lyria-3-pro-preview",
  contents: "An epic cinematic orchestral piece about a journey home. " +
            "Starts with a solo piano intro, builds through sweeping " +
            "strings, and climaxes with a massive wall of sound.",
  config: config
);

Analisar a resposta

A resposta do Lyria 3 contém várias partes. As partes de texto contêm a letra gerada ou uma descrição JSON da estrutura da música. As partes com inline_data contêm os bytes de áudio.

Python

lyrics = []
audio_data = None

for part in response.parts:
    if part.text is not None:
        lyrics.append(part.text)
    elif part.inline_data is not None:
        audio_data = part.inline_data.data

if lyrics:
    print("Lyrics:\n" + "\n".join(lyrics))

if audio_data:
    with open("output.mp3", "wb") as f:
        f.write(audio_data)

JavaScript

const lyrics = [];
let audioData = null;

for (const part of response.candidates[0].content.parts) {
  if (part.text) {
    lyrics.push(part.text);
  } else if (part.inlineData) {
    audioData = Buffer.from(part.inlineData.data, "base64");
  }
}

if (lyrics.length) {
  console.log("Lyrics:\n" + lyrics.join("\n"));
}

if (audioData) {
  fs.writeFileSync("output.mp3", audioData);
}

Go

var lyrics []string
var audioData []byte

for _, part := range result.Candidates[0].Content.Parts {
    if part.Text != "" {
        lyrics = append(lyrics, part.Text)
    } else if part.InlineData != nil {
        audioData = part.InlineData.Data
    }
}

if len(lyrics) > 0 {
    fmt.Println("Lyrics:\n" + strings.Join(lyrics, "\n"))
}

if audioData != nil {
    err := os.WriteFile("output.mp3", audioData, 0644)
    if err != nil {
        log.Fatal(err)
    }
}

Java

List<String> lyrics = new ArrayList<>();
byte[] audioData = null;

for (Part part : response.parts()) {
  if (part.text().isPresent()) {
    lyrics.add(part.text().get());
  } else if (part.inlineData().isPresent()) {
    audioData = part.inlineData().get().data().get();
  }
}

if (!lyrics.isEmpty()) {
  System.out.println("Lyrics:\n" + String.join("\n", lyrics));
}

if (audioData != null) {
  Files.write(Paths.get("output.mp3"), audioData);
}

C#

var lyrics = new List<string>();
byte[] audioData = null;

foreach (var part in response.Candidates[0].Content.Parts) {
  if (part.Text != null) {
    lyrics.Add(part.Text);
  } else if (part.InlineData != null) {
    audioData = part.InlineData.Data;
  }
}

if (lyrics.Count > 0) {
  Console.WriteLine("Lyrics:\n" + string.Join("\n", lyrics));
}

if (audioData != null) {
  await File.WriteAllBytesAsync("output.mp3", audioData);
}

REST

# The output from the REST API is a JSON object containing base64 encoded data.
# You can extract the text or the audio data using a tool like jq.
# To extract the audio and save it to a file:
curl ... | jq -r '.candidates[0].content.parts[] | select(.inlineData) | .inlineData.data' | base64 -d > output.mp3

Gerar música com base em imagens

O Lyria 3 aceita entradas multimodais. É possível fornecer até 10 imagens junto com o comando de texto, e o modelo vai compor músicas inspiradas no conteúdo visual.

Python

from PIL import Image

image = Image.open("desert_sunset.jpg")

response = client.models.generate_content(
    model="lyria-3-pro-preview",
    contents=[
        "An atmospheric ambient track inspired by the mood and "
        "colors in this image.",
        image,
    ],
    config=types.GenerateContentConfig(
        response_modalities=["AUDIO", "TEXT"],
    ),
)

JavaScript

const imageData = fs.readFileSync("desert_sunset.jpg");
const base64Image = imageData.toString("base64");

const response = await ai.models.generateContent({
  model: "lyria-3-pro-preview",
  contents: [
    { text: "An atmospheric ambient track inspired by the mood " +
            "and colors in this image." },
    {
      inlineData: {
        mimeType: "image/jpeg",
        data: base64Image,
      },
    },
  ],
  config: {
    responseModalities: ["AUDIO", "TEXT"],
  },
});

Go

imgData, err := os.ReadFile("desert_sunset.jpg")
if err != nil {
    log.Fatal(err)
}

parts := []*genai.Part{
    genai.NewPartFromText("An atmospheric ambient track inspired " +
        "by the mood and colors in this image."),
    &genai.Part{
        InlineData: &genai.Blob{
            MIMEType: "image/jpeg",
            Data:     imgData,
        },
    },
}

contents := []*genai.Content{
    genai.NewContentFromParts(parts, genai.RoleUser),
}

result, err := client.Models.GenerateContent(
    ctx,
    "lyria-3-pro-preview",
    contents,
    config,
)

Java

GenerateContentResponse response = client.models.generateContent(
    "lyria-3-pro-preview",
    Content.fromParts(
        Part.fromText("An atmospheric ambient track inspired by "
            + "the mood and colors in this image."),
        Part.fromBytes(
            Files.readAllBytes(Path.of("desert_sunset.jpg")),
            "image/jpeg")),
    config);

REST

curl -s -X POST \
  "https://generativelanguage.googleapis.com/v1beta/models/lyria-3-pro-preview:generateContent" \
  -H "x-goog-api-key: $GEMINI_API_KEY" \
  -H 'Content-Type: application/json' \
  -d "{
    \"contents\": [{
      \"parts\":[
          {\"text\": \"An atmospheric ambient track inspired by the mood and colors in this image.\"},
          {
            \"inline_data\": {
              \"mime_type\":\"image/jpeg\",
              \"data\": \"<BASE64_IMAGE_DATA>\"
            }
          }
      ]
    }],
    \"generationConfig\": {
      \"responseModalities\": [\"AUDIO\", \"TEXT\"]
    }
  }"

C#

var response = await client.Models.GenerateContentAsync(
  model: "lyria-3-pro-preview",
  contents: new List<Part> {
    Part.FromText("An atmospheric ambient track inspired by the mood and colors in this image."),
    Part.FromBytes(await File.ReadAllBytesAsync("desert_sunset.jpg"), "image/jpeg")
  },
  config: config
);

Fornecer letras personalizadas

É possível escrever sua própria letra e incluí-la no comando. Use tags de seção como [Verse], [Chorus] e [Bridge] para ajudar o modelo a entender a estrutura da música:

Python

prompt = """
Create a dreamy indie pop song with the following lyrics:

[Verse 1]
Walking through the neon glow,
city lights reflect below,
every shadow tells a story,
every corner, fading glory.

[Chorus]
We are the echoes in the night,
burning brighter than the light,
hold on tight, don't let me go,
we are the echoes down below.

[Verse 2]
Footsteps lost on empty streets,
rhythms sync to heartbeats,
whispers carried by the breeze,
dancing through the autumn leaves.
"""

response = client.models.generate_content(
    model="lyria-3-pro-preview",
    contents=prompt,
    config=types.GenerateContentConfig(
        response_modalities=["AUDIO", "TEXT"],
    ),
)

JavaScript

const prompt = `
Create a dreamy indie pop song with the following lyrics:

[Verse 1]
Walking through the neon glow,
city lights reflect below,
every shadow tells a story,
every corner, fading glory.

[Chorus]
We are the echoes in the night,
burning brighter than the light,
hold on tight, don't let me go,
we are the echoes down below.

[Verse 2]
Footsteps lost on empty streets,
rhythms sync to heartbeats,
whispers carried by the breeze,
dancing through the autumn leaves.
`;

const response = await ai.models.generateContent({
  model: "lyria-3-pro-preview",
  contents: prompt,
  config: {
    responseModalities: ["AUDIO", "TEXT"],
  },
});

Go

prompt := `
Create a dreamy indie pop song with the following lyrics:

[Verse 1]
Walking through the neon glow,
city lights reflect below,
every shadow tells a story,
every corner, fading glory.

[Chorus]
We are the echoes in the night,
burning brighter than the light,
hold on tight, don't let me go,
we are the echoes down below.

[Verse 2]
Footsteps lost on empty streets,
rhythms sync to heartbeats,
whispers carried by the breeze,
dancing through the autumn leaves.
`

result, err := client.Models.GenerateContent(
    ctx,
    "lyria-3-pro-preview",
    genai.Text(prompt),
    config,
)

Java

String prompt = """
    Create a dreamy indie pop song with the following lyrics:

    [Verse 1]
    Walking through the neon glow,
    city lights reflect below,
    every shadow tells a story,
    every corner, fading glory.

    [Chorus]
    We are the echoes in the night,
    burning brighter than the light,
    hold on tight, don't let me go,
    we are the echoes down below.

    [Verse 2]
    Footsteps lost on empty streets,
    rhythms sync to heartbeats,
    whispers carried by the breeze,
    dancing through the autumn leaves.
    """;

GenerateContentResponse response = client.models.generateContent(
    "lyria-3-pro-preview",
    prompt,
    config);

C#

var prompt = @"
Create a dreamy indie pop song with the following lyrics:

[Verse 1]
Walking through the neon glow,
city lights reflect below,
every shadow tells a story,
every corner, fading glory.

[Chorus]
We are the echoes in the night,
burning brighter than the light,
hold on tight, don't let me go,
we are the echoes down below.

[Verse 2]
Footsteps lost on empty streets,
rhythms sync to heartbeats,
whispers carried by the breeze,
dancing through the autumn leaves.
";

var response = await client.Models.GenerateContentAsync(
  model: "lyria-3-pro-preview",
  contents: prompt,
  config: config
);

REST

curl -s -X POST \
  "https://generativelanguage.googleapis.com/v1beta/models/lyria-3-pro-preview:generateContent" \
  -H "x-goog-api-key: $GEMINI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [{
      "parts": [
        {"text": "Create a dreamy indie pop song with the following lyrics: ..."}
      ]
    }],
    "generationConfig": {
      "responseModalities": ["AUDIO", "TEXT"]
    }
  }'

Controlar o tempo e a estrutura

É possível especificar exatamente o que acontece em momentos específicos da música usando marcações de tempo. Isso é útil para controlar quando os instrumentos entram, quando as letras são entregues e como a música progride:

Python

prompt = """
[0:00 - 0:10] Intro: Begin with a soft lo-fi beat and muffled
              vinyl crackle.
[0:10 - 0:30] Verse 1: Add a warm Fender Rhodes piano melody
              and gentle vocals singing about a rainy morning.
[0:30 - 0:50] Chorus: Full band with upbeat drums and soaring
              synth leads. The lyrics are hopeful and uplifting.
[0:50 - 1:00] Outro: Fade out with the piano melody alone.
"""

response = client.models.generate_content(
    model="lyria-3-pro-preview",
    contents=prompt,
    config=types.GenerateContentConfig(
        response_modalities=["AUDIO", "TEXT"],
    ),
)

JavaScript

const prompt = `
[0:00 - 0:10] Intro: Begin with a soft lo-fi beat and muffled
              vinyl crackle.
[0:10 - 0:30] Verse 1: Add a warm Fender Rhodes piano melody
              and gentle vocals singing about a rainy morning.
[0:30 - 0:50] Chorus: Full band with upbeat drums and soaring
              synth leads. The lyrics are hopeful and uplifting.
[0:50 - 1:00] Outro: Fade out with the piano melody alone.
`;

const response = await ai.models.generateContent({
  model: "lyria-3-pro-preview",
  contents: prompt,
  config: {
    responseModalities: ["AUDIO", "TEXT"],
  },
});

Go

prompt := `
[0:00 - 0:10] Intro: Begin with a soft lo-fi beat and muffled
              vinyl crackle.
[0:10 - 0:30] Verse 1: Add a warm Fender Rhodes piano melody
              and gentle vocals singing about a rainy morning.
[0:30 - 0:50] Chorus: Full band with upbeat drums and soaring
              synth leads. The lyrics are hopeful and uplifting.
[0:50 - 1:00] Outro: Fade out with the piano melody alone.
`

result, err := client.Models.GenerateContent(
    ctx,
    "lyria-3-pro-preview",
    genai.Text(prompt),
    config,
)

Java

String prompt = """
    [0:00 - 0:10] Intro: Begin with a soft lo-fi beat and muffled
                  vinyl crackle.
    [0:10 - 0:30] Verse 1: Add a warm Fender Rhodes piano melody
                  and gentle vocals singing about a rainy morning.
    [0:30 - 0:50] Chorus: Full band with upbeat drums and soaring
                  synth leads. The lyrics are hopeful and uplifting.
    [0:50 - 1:00] Outro: Fade out with the piano melody alone.
    """;

GenerateContentResponse response = client.models.generateContent(
    "lyria-3-pro-preview",
    prompt,
    config);

C#

var prompt = @"
[0:00 - 0:10] Intro: Begin with a soft lo-fi beat and muffled
              vinyl crackle.
[0:10 - 0:30] Verse 1: Add a warm Fender Rhodes piano melody
              and gentle vocals singing about a rainy morning.
[0:30 - 0:50] Chorus: Full band with upbeat drums and soaring
              synth leads. The lyrics are hopeful and uplifting.
[0:50 - 1:00] Outro: Fade out with the piano melody alone.
";

var response = await client.Models.GenerateContentAsync(
  model: "lyria-3-pro-preview",
  contents: prompt,
  config: config
);

REST

curl -s -X POST \
  "https://generativelanguage.googleapis.com/v1beta/models/lyria-3-pro-preview:generateContent" \
  -H "x-goog-api-key: $GEMINI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [{
      "parts": [
        {"text": "[0:00 - 0:10] Intro: ..."}
      ]
    }],
    "generationConfig": {
      "responseModalities": ["AUDIO", "TEXT"]
    }
  }'

Gerar faixas instrumentais

Para música de fundo, trilhas sonoras de jogos ou qualquer caso de uso em que os vocais não sejam necessários, é possível solicitar que o modelo produza faixas apenas instrumentais:

Python

response = client.models.generate_content(
    model="lyria-3-clip-preview",
    contents="A bright chiptune melody in C Major, retro 8-bit "
             "video game style. Instrumental only, no vocals.",
    config=types.GenerateContentConfig(
        response_modalities=["AUDIO", "TEXT"],
    ),
)

JavaScript

const response = await ai.models.generateContent({
  model: "lyria-3-clip-preview",
  contents: "A bright chiptune melody in C Major, retro 8-bit " +
            "video game style. Instrumental only, no vocals.",
  config: {
    responseModalities: ["AUDIO", "TEXT"],
  },
});

Go

result, err := client.Models.GenerateContent(
    ctx,
    "lyria-3-clip-preview",
    genai.Text("A bright chiptune melody in C Major, retro 8-bit " +
               "video game style. Instrumental only, no vocals."),
    config,
)

Java

GenerateContentResponse response = client.models.generateContent(
    "lyria-3-clip-preview",
    "A bright chiptune melody in C Major, retro 8-bit "
        + "video game style. Instrumental only, no vocals.",
    config);

C#

var response = await client.Models.GenerateContentAsync(
  model: "lyria-3-clip-preview",
  contents: "A bright chiptune melody in C Major, retro 8-bit " +
            "video game style. Instrumental only, no vocals.",
  config: config
);

REST

curl -s -X POST \
  "https://generativelanguage.googleapis.com/v1beta/models/lyria-3-clip-preview:generateContent" \
  -H "x-goog-api-key: $GEMINI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [{
      "parts": [
        {"text": "A bright chiptune melody in C Major, retro 8-bit video game style. Instrumental only, no vocals."}
      ]
    }],
    "generationConfig": {
      "responseModalities": ["AUDIO", "TEXT"]
    }
  }'

Gerar música em idiomas diferentes

O Lyria 3 gera letras no idioma do comando. Para gerar uma música com letras em francês, escreva o comando nesse idioma. O modelo adapta o estilo vocal e a pronúncia para corresponder ao idioma.

Python

response = client.models.generate_content(
    model="lyria-3-pro-preview",
    contents="Crée une chanson pop romantique en français sur un "
             "coucher de soleil à Paris. Utilise du piano et de "
             "la guitare acoustique.",
    config=types.GenerateContentConfig(
        response_modalities=["AUDIO", "TEXT"],
    ),
)

JavaScript

const response = await ai.models.generateContent({
  model: "lyria-3-pro-preview",
  contents: "Crée une chanson pop romantique en français sur un " +
            "coucher de soleil à Paris. Utilise du piano et de " +
            "la guitare acoustique.",
  config: {
    responseModalities: ["AUDIO", "TEXT"],
  },
});

Go

result, err := client.Models.GenerateContent(
    ctx,
    "lyria-3-pro-preview",
    genai.Text("Crée une chanson pop romantique en français sur un " +
               "coucher de soleil à Paris. Utilise du piano et de " +
               "la guitare acoustique."),
    config,
)

Java

GenerateContentResponse response = client.models.generateContent(
    "lyria-3-pro-preview",
    "Crée une chanson pop romantique en français sur un "
        + "coucher de soleil à Paris. Utilise du piano et de "
        + "la guitare acoustique.",
    config);

C#

var response = await client.Models.GenerateContentAsync(
  model: "lyria-3-pro-preview",
  contents: "Crée une chanson pop romantique en français sur un " +
            "coucher de soleil à Paris. Utilise du piano et de " +
            "la guitare acoustique.",
  config: config
);

REST

curl -s -X POST \
  "https://generativelanguage.googleapis.com/v1beta/models/lyria-3-pro-preview:generateContent" \
  -H "x-goog-api-key: $GEMINI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [{
      "parts": [
        {"text": "Crée une chanson pop romantique en français sur un coucher de soleil à Paris. Utilise du piano et de la guitare acoustique."}
      ]
    }],
    "generationConfig": {
      "responseModalities": ["AUDIO", "TEXT"]
    }
  }'

Inteligência do modelo

O Lyria 3 analisa o processo de comando em que o modelo raciocina sobre a estrutura musical (introdução, verso, refrão, ponte etc.) com base no comando. Isso acontece antes da geração do áudio e garante a coerência estrutural e a musicalidade.

API Interactions

É possível usar modelos do Lyria 3 com a API Interactions; uma interface unificada para interagir com modelos e agentes do Gemini. Ela simplifica o gerenciamento de estado e tarefas de longa duração para casos de uso multimodais complexos.

Python

from google import genai

client = genai.Client()

interaction = client.interactions.create(
    model="lyria-3-pro-preview",
    input="An epic cinematic orchestral piece about a journey home. " +
          "Starts with a solo piano intro, builds through sweeping " +
          "strings, and climaxes with a massive wall of sound.",
    response_modalities=["AUDIO", "TEXT"]
)

for output in interaction.outputs:
    if output.text:
        print(output.text)
    elif output.inline_data:
         with open("interaction_output.mp3", "wb") as f:
            f.write(output.inline_data.data)
         print("Audio saved to interaction_output.mp3")

JavaScript

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

const client = new GoogleGenAI({});

const interaction = await client.interactions.create({
  model: 'lyria-3-pro-preview',
  input: 'An epic cinematic orchestral piece about a journey home. ' +
         'Starts with a solo piano intro, builds through sweeping ' +
         'strings, and climaxes with a massive wall of sound.',
  responseModalities: ['AUDIO', 'TEXT'],
});

for (const output of interaction.outputs) {
  if (output.text) {
    console.log(output.text);
  } else if (output.inlineData) {
    const buffer = Buffer.from(output.inlineData.data, 'base64');
    fs.writeFileSync('interaction_output.mp3', buffer);
    console.log('Audio saved to interaction_output.mp3');
  }
}

REST

curl -X POST "https://generativelanguage.googleapis.com/v1beta/interactions" \
-H "Content-Type: application/json" \
-H "x-goog-api-key: $GEMINI_API_KEY" \
-d '{
    "model": "lyria-3-pro-preview",
    "input": "An epic cinematic orchestral piece about a journey home. Starts with a solo piano intro, builds through sweeping strings, and climaxes with a massive wall of sound.",
    "responseModalities": ["AUDIO", "TEXT"]
}'

Guia para a criação de comandos

Quanto mais específico for o comando, melhores serão os resultados. Confira o que você pode incluir para orientar a geração:

  • Gênero: especifique um gênero ou uma mistura de gêneros (por exemplo, "lo-fi hip hop", "jazz fusion", "orquestra cinematográfica").
  • Instrumentos: nomeie instrumentos específicos (por exemplo, "piano Fender Rhodes", "guitarra slide", "bateria eletrônica TR-808").
  • BPM: defina o tempo (por exemplo, "120 BPM", "tempo lento em torno de 70 BPM").
  • Tom/escala: especifique um tom musical (por exemplo, "em sol maior", "ré menor").
  • Clima e atmosfera: use adjetivos descritivos (por exemplo, "nostálgico", "agressivo", "etéreo", "sonhador").
  • Estrutura: use tags como [Verse], [Chorus], [Bridge], [Intro], [Outro] ou marcações de tempo para controlar a progressão da música.
  • Duração: o modelo Clip sempre produz clipes de 30 segundos. Para o modelo Pro, especifique a duração desejada no comando (por exemplo, "criar uma música de 2 minutos") ou use marcações de tempo para controlar a duração.

Exemplos de comandos

Confira alguns exemplos de comandos eficazes:

  • "A 30-second lofi hip hop beat with dusty vinyl crackle, mellow Rhodes piano chords, a slow boom-bap drum pattern at 85 BPM, and a jazzy upright bass line. Instrumental only."
  • "An upbeat, feel-good pop song in G major at 120 BPM with bright acoustic guitar strumming, claps, and warm vocal harmonies about a summer road trip."
  • "A dark, atmospheric trap beat at 140 BPM with heavy 808 bass, eerie synth pads, sharp hi-hats, and a haunting vocal sample. In D minor."

Práticas recomendadas

  • Itere primeiro com o Clip. Use o modelo lyria-3-clip-preview mais rápido para testar comandos antes de se comprometer com uma geração completa com lyria-3-pro-preview.
  • Faça uma descrição específica. Comandos vagos produzem resultados genéricos. Mencione instrumentos, BPM, tom, clima e estrutura para a melhor saída.
  • Combine o idioma. Crie o comando no idioma em que você quer a letra.
  • Use tags de seção. As tags [Verse], [Chorus] e [Bridge] oferecem ao modelo uma estrutura clara a seguir.
  • Separe a letra das instruções. Ao fornecer letras personalizadas, separe-as claramente das instruções de direção musical.

Limitações

  • Segurança: todos os comandos são verificados por filtros de segurança. Os comandos que acionam os filtros serão bloqueados. Isso inclui comandos que solicitam vozes de artistas específicos ou a geração de letras protegidas por direitos autorais.
  • Marca-d'água: todo o áudio gerado inclui uma marca-d'água digital do SynthID para identificação. Essa marca-d'água é imperceptível ao ouvido humano e não afeta a experiência de audição.
  • Edição de várias rodadas: a geração de música é um processo de rodada única. A edição iterativa ou o refinamento de um clipe gerado por vários comandos não é aceito na versão atual do Lyria 3.
  • Duração: o modelo Clip sempre gera clipes de 30 segundos. O modelo Pro gera músicas que duram alguns minutos. A duração exata pode ser influenciada pelo comando.
  • Determinismo: os resultados podem variar entre chamadas, mesmo com o mesmo comando.

A seguir