MusicGen LTB

MusicGen LTB is an advanced model for text-to-music generation, offering audio continuation and melody mimicking. Try it now and see the results!

Platform: Replicate

Music GenerationAudio ContinuationMelody Mimicking

49 runs

L40S

License Check Required

🚀Function Overview

Generates music based on text descriptions and audio inputs, with options for continuation or melody mimicry.

Key Features

Text-to-music generation using prompts
Audio file input for melody continuation or imitation
Adjustable audio duration and generation parameters
Control over diversity and output adherence via temperature and guidance settings
Support for multi-band diffusion decoding

Use Cases

•Creating original music tracks from text descriptions
•Extending existing music compositions via audio continuation
•Generating music that mimics specific melodies for remixing
•Producing background music for videos or games

⚙️Input Parameters

prompt

string

A description of the music you want to generate.

input_audio

string

An audio file that will influence the generated music. If `continuation` is `True`, the generated music will be a continuation of the audio file. Otherwise, the generated music will mimic the audio file's melody.

duration

integer

Duration of the generated audio in seconds.

continuation

boolean

If `True`, generated music will continue `melody`. Otherwise, generated music will mimic `audio_input`'s melody.

continuation_start

integer

Start time of the audio file to use for continuation.

continuation_end

integer

End time of the audio file to use for continuation. If -1 or None, will default to the end of the audio clip.

multi_band_diffusion

boolean

If `True`, the EnCodec tokens will be decoded with MultiBand Diffusion. Only works with non-stereo models.

normalization_strategy

string

Strategy for normalizing audio.

top_k

integer

Reduces sampling to the k most likely tokens.

top_p

number

Reduces sampling to tokens with cumulative probability of p. When set to `0` (default), top_k sampling is used.

temperature

number

Controls the 'conservativeness' of the sampling process. Higher temperature means more diversity.

classifier_free_guidance

integer

Increases the influence of inputs on the output. Higher values produce lower-varience outputs that adhere more closely to inputs.

output_format

string

Output format for generated audio.

seed

integer

Seed for random number generator. If None or -1, a random seed will be used.

💡Usage Examples

Example 1

Input Parameters

{
  "top_k": 250,
  "top_p": 0,
  "prompt": " Driving 80s-style electronic with lush synth textures, arpeggiated basslines, retro guitar licks, and dramatic gated drums. Italo disco energy with a moody, nostalgic atmosphere. BPM: 98",
  "duration": 8,
  "temperature": 1,
  "continuation": false,
  "output_format": "wav",
  "continuation_start": 0,
  "multi_band_diffusion": false,
  "normalization_strategy": "loudness",
  "classifier_free_guidance": 3
}

Output Results

https://replicate.delivery/xezq/VE3pgq1rfHyHY67kl7gMHIGwhlAwBkeHnQPQR0y6zxzrS5wUA/out.wav

Quick Actions

Use NowView Documentation

Technical Specifications

Hardware Type: L40S
Run Count: 49
Commercial Use: Unknown/Restricted
Platform: Replicate

Related Keywords

Text-to-Music GenerationAudio ContinuationMelody MimickingOriginal Music CreationBackground Music GenerationExtend Compositions

Related Models

DrumTest2 Rhythmic Audio Transformer

Transforms any rhythmic sound—a drum kit, beatboxing, a toy drum, even drumming on your belly—into a pro-quality performance on Zohar's studio drum kit.

Speaker Diarization

Speaker Diarization with "pyannote/speaker-diarization-3.1"

Resemble Enhance AI

Optimizes audio files with speech