turian/multi-task-music-transcription 🖼️❓ → 🖼️

▶️ 165 runs 📅 Jul 2024 ⚙️ Cog 0.9.13 🔗 GitHub 📄 Paper ⚖️ License

audio-to-midi music-transcription music-understanding

About

MT3: Multi-Task Multitrack Music Transcription

Example Output

Output

Performance Metrics

46.74s Prediction Time

106.70s Total Time

All Input Parameters

{
  "audio_file": "https://replicate.delivery/pbxt/LwqzMVvrrQGdaOgpRrwpLYaNdEoAfQBCvB8mxptn0aRqF6nH/762258__lolamoore__energetic-indie-rock-track-with-uplifting-vibes.mp3",
  "model_type": "mt3"
}

Input Parameters

audio_file (required) Type: string: Input audio file
model_type Default: mt3: Model type: ismir2021 for piano, mt3 for multi-instrument

Output Schema

Output

Type: string • Format: uri

Example Execution Logs

2024-11-10 13:40:23.646358: I tensorflow/core/common_runtime/gpu/gpu_device.cc:2021] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 79091 MB memory:  -> device: 0, name: NVIDIA A100-SXM4-80GB, pci bus id: 0000:57:00.0, compute capability: 8.0
WARNING:tensorflow:From /root/.pyenv/versions/3.10.14/lib/python3.10/site-packages/tensorflow/python/util/nest_util.py:1105: _RandomDataset.__init__ (from tensorflow.python.data.experimental.ops.random_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use `tf.data.Dataset.random(...)`.
WARNING:tensorflow:From /root/.pyenv/versions/3.10.14/lib/python3.10/site-packages/tensorflow/python/util/nest_util.py:1105: _RandomDataset.__init__ (from tensorflow.python.data.experimental.ops.random_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use `tf.data.Dataset.random(...)`.
2024-11-10 13:40:58.294195: I tensorflow/core/framework/local_rendezvous.cc:404] Local rendezvous is aborting with status: OUT_OF_RANGE: End of sequence

Version Details

Version ID: 159965ec81a23b88d50e4636f9539f2cbb6f50503d8f14e915069526b4a9edfb
Version Created: July 31, 2024

Run on Replicate →