Input

Run this model in Node.js with one line of code:

npx create-replicate --model=mtg/essentia-bpm

or set up a project from scratch

Install Replicate’s Node.js client library:

npm install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run mtg/essentia-bpm using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "mtg/essentia-bpm:b3045c359817fea53678791886d50aa3e3a995dc4796fe74db0de156d5074a43",
  {
    input: {
      url: "https://www.youtube.com/watch?v=5-XfBN3olvE",
      algo_type: "deepsquare-k16"
    }
  }
);
console.log(output);

To learn more, take a look at the guide on getting started with Node.js.

Install Replicate’s Python client library:

pip install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import the client:

import replicate

Run mtg/essentia-bpm using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "mtg/essentia-bpm:b3045c359817fea53678791886d50aa3e3a995dc4796fe74db0de156d5074a43",
    input={
        "url": "https://www.youtube.com/watch?v=5-XfBN3olvE",
        "algo_type": "deepsquare-k16"
    }
)
print(output)

To learn more, take a look at the guide on getting started with Python.

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Run mtg/essentia-bpm using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "b3045c359817fea53678791886d50aa3e3a995dc4796fe74db0de156d5074a43",
    "input": {
      "url": "https://www.youtube.com/watch?v=5-XfBN3olvE",
      "algo_type": "deepsquare-k16"
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

Output

Rendering markdown...

{
  "completed_at": "2023-11-30T10:58:18.095332Z",
  "created_at": "2023-11-30T10:58:07.874800Z",
  "data_removed": false,
  "error": null,
  "id": "xmdbg2rbdv6u2vd2zlynkq73uu",
  "input": {
    "url": "https://www.youtube.com/watch?v=5-XfBN3olvE",
    "algo_type": "deepsquare-k16"
  },
  "logs": "[youtube] 5-XfBN3olvE: Downloading webpage\n[youtube] 5-XfBN3olvE: Downloading player 5753e790\n[dashsegments] Total fragments: 1\n[download] Destination: /tmp/tmpv7ejeaw3/audio.webm\n[download]   0.0% of ~4.95MiB at  6.94KiB/s ETA 12:10\n[download]   0.1% of ~4.95MiB at 20.79KiB/s ETA 04:03\n[download]   0.1% of ~4.95MiB at 48.42KiB/s ETA 01:44\n[download]   0.3% of ~4.95MiB at 103.49KiB/s ETA 00:48\n[download]   0.6% of ~4.95MiB at 201.98KiB/s ETA 00:24\n[download]   1.2% of ~4.95MiB at 388.21KiB/s ETA 00:12\n[download]   2.5% of ~4.95MiB at 738.33KiB/s ETA 00:06\n[download]   5.0% of ~4.95MiB at  1.37MiB/s ETA 00:03\n[download]  10.1% of ~4.95MiB at  2.59MiB/s ETA 00:01\n[download]  20.2% of ~4.95MiB at  4.92MiB/s ETA 00:00\n[download]  40.4% of ~4.95MiB at  9.35MiB/s ETA 00:00\n[download]  80.8% of ~4.95MiB at 17.58MiB/s ETA 00:00\n[download] 100.0% of ~4.95MiB at 21.41MiB/s ETA 00:00\n[download] 100.0% of ~4.95MiB at 21.36MiB/s ETA 00:00\n[download] 100% of 4.95MiB in 00:00\n[ffmpeg] Destination: /tmp/tmpv7ejeaw3/audio.wav\nDeleting original file /tmp/tmpv7ejeaw3/audio.webm (pass -k to keep)\nloading audio...\nrunning analysis algorithm...\nDEBUG size before resample 14819841\nDEBUG size after resample 3704961\nDEBUG [0. 0. 0. ... 0. 0. 0.]\n2023-11-30 10:58:17.387081: I tensorflow/core/platform/profile_utils/cpu_utils.cc:114] CPU Frequency: 2299995000 Hz\ndone!",
  "metrics": {
    "predict_time": 8.331015,
    "total_time": 10.220532
  },
  "output": "https://replicate.delivery/pbxt/jitCG30GoS6LPVBbIxsLJVmc4k25bDZgFSc7KPIA2fmk5weRA/out.md",
  "started_at": "2023-11-30T10:58:09.764317Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/xmdbg2rbdv6u2vd2zlynkq73uu",
    "cancel": "https://api.replicate.com/v1/predictions/xmdbg2rbdv6u2vd2zlynkq73uu/cancel"
  },
  "version": "b3045c359817fea53678791886d50aa3e3a995dc4796fe74db0de156d5074a43"
}

Generated in

8.3 seconds

Tweak itReport

Run time and cost

This model costs approximately $0.00053 to run on Replicate, or 1886 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on CPU hardware. Predictions typically complete within 6 seconds. The predict time for this model varies significantly based on the inputs.

Readme

Tempo BPM estimation with Essentia

This demo showcases tempo estimation algorithms available in Essentia:

degara: https://essentia.upf.edu/reference/std_RhythmExtractor2013.html
multifeature: https://essentia.upf.edu/reference/std_RhythmExtractor2013.html
percival: https://essentia.upf.edu/reference/std_PercivalBpmEstimator.html
deepsquare-k16: https://essentia.upf.edu/models.html#tempocnn

License

These algorithms are part of Essentia made by MTG-UPF and are available under Affero GPLv3 license and commercial license.