1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
import requests
import base64
# Use this function to convert an image file from the filesystem to base64
def image_file_to_base64(image_path):
with open(image_path, 'rb') as f:
image_data = f.read()
return base64.b64encode(image_data).decode('utf-8')
# Use this function to fetch an image from a URL and convert it to base64
def image_url_to_base64(image_url):
response = requests.get(image_url)
image_data = response.content
return base64.b64encode(image_data).decode('utf-8')
api_key = "YOUR_API_KEY"
url = "https://api.segmind.com/v1/meta-musicgen-medium"
# Request payload
data = {
"prompt": "lo-fi music with a soothing melody",
"duration": 10,
"seed": 42
}
headers = {'x-api-key': api_key}
response = requests.post(url, json=data, headers=headers)
print(response.content) # The response is the generated imageAttributes
prompt to generate music
duration of the audio
min : 1,
max : 30
Seed for audio generation.
min : -1,
max : 9999
To keep track of your credit usage, you can inspect the response headers of each API call. The x-remaining-credits property will indicate the number of remaining credits in your account. Ensure you monitor this value to avoid any disruptions in your API usage.
MusicGen: Transforming Text into Music
MusicGen by Meta is an advanced text-to-music model designed to generate high-quality music samples from text descriptions or audio prompts. Leveraging a single-stage auto-regressive Transformer architecture, MusicGen is trained on a 32kHz EnCodec tokenizer with four codebooks sampled at 50 Hz. This innovative approach allows for efficient and high-fidelity music generation.
Key Features of MusicGen
-
Text-to-Music Generation: Converts textual descriptions into diverse and high-quality music samples.
-
Auto-Regressive Transformer: Utilizes a single-stage auto-regressive Transformer model for seamless music generation.
-
Efficient Training: Trained on a 32kHz EnCodec tokenizer with four codebooks, enabling efficient processing and high-quality output.
-
Parallel Prediction: Introduces a small delay between codebooks, allowing parallel prediction and reducing the number of auto-regressive steps to 50 per second of audio.
Use cases
-
Music Production: Generate unique music tracks based on textual descriptions for use in various media.
-
Creative Projects: Enhance creative projects with custom-generated music that matches specific themes or moods.
-
Interactive Experiences: Integrate into interactive applications to provide dynamic and responsive musical experiences.
Popular Models
