sakemin/musicongen 🔢📝❓ → 🖼️

▶️ 168 runs 📅 Aug 2024 ⚙️ Cog 0.8.6 🔗 GitHub 📄 Paper ⚖️ License
chord-conditioning music-generation text-to-music

About

"MusiConGen: Rhythm and chord control for Transformer-based text-to-music generation"

Example Output

Prompt:

"A laid-back blues shuffle with a relaxed tempo, warm guitar tones, and a comfortable groove, perfect for a slow dance or a night in. Instruments: electric guitar, bass, drums."

Output

Example output

Performance Metrics

48.22s Prediction Time
175.86s Total Time
All Input Parameters 
{
  "bpm": 120,
  "top_k": 250,
  "top_p": 0,
  "prompt": "A laid-back blues shuffle with a relaxed tempo, warm guitar tones, and a comfortable groove, perfect for a slow dance or a night in. Instruments: electric guitar, bass, drums.",
  "duration": 30,
  "time_sig": "4/4",
  "temperature": 1,
  "text_chords": "C G A:min F",
  "output_format": "wav",
  "classifier_free_guidance": 3
}
Input Parameters
bpm Type: numberDefault: 120
BPM condition for the generated output. Chord and rhythm conditions are generated upon this value. This will be appended at the end of `prompt`.
seed Type: integer
Seed for random number generator. If `None` or `-1`, a random seed will be used.
top_k Type: integerDefault: 250
Reduces sampling to the k most likely tokens.
top_p Type: numberDefault: 0
Reduces sampling to tokens with cumulative probability of p. When set to `0` (default), top_k sampling is used.
prompt Type: stringDefault: A laid-back blues shuffle with a relaxed tempo, warm guitar tones, and a comfortable groove, perfect for a slow dance or a night in. Instruments: electric guitar, bass, drums.
A description of the music you want to generate.
duration Type: integerDefault: 30Range: ∞ - 30
Duration of the generated audio in seconds.
time_sig Type: stringDefault: 4/4
Meter value for the generate output. Chord and rhythm conditions are generated upon this value. This will be appended at the end of `prompt`.
temperature Type: numberDefault: 1
Controls the 'conservativeness' of the sampling process. Higher temperature means more diversity.
text_chords Type: stringDefault: C G A:min F
A text based chord progression condition. Single uppercase alphabet character(eg. `C`) is considered as a major chord. Chord attributes like(`maj`, `min`, `dim`, `aug`, `min6`, `maj6`, `min7`, `minmaj7`, `maj7`, `7`, `dim7`, `hdim7`, `sus2` and `sus4`) can be added to the root alphabet character after `:`.(eg. `A:min7`) Each chord token splitted by `SPACE` is allocated to a single bar. If more than one chord must be allocated to a single bar, cluster the chords adding with `,` without any `SPACE`.(eg. `C,C:7 G, E:min A:min`)
output_format Default: wav
Output format for generated audio.
classifier_free_guidance Type: integerDefault: 3
Increases the influence of inputs on the output. Higher values produce lower-varience outputs that adhere more closely to inputs.
Output Schema

Output

Type: string • Format: uri

Example Execution Logs 
Using seed 2360678342
Version Details
Version ID
a05ec8bdf5cc902cd849077d985029ce9b05e3dfb98a2d74accc9c94fdf15747
Version Created
August 3, 2024
Run on Replicate →