SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2209.14458 · 2022

The Chamber Ensemble Generator: Limitless High-Quality MIR Data via Generative Modeling

arXiv · PDF · Open in the Atlas

Code that ran

8 of 9 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 7 have no contract verdict; 1 failure of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2209.14458.json · badge: /api/ran/2209.14458.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
lukewys/chamber-ensemble-generator canonical (classifier label)
classifier basis not recorded on the link
8 ran · 1 no recorded run
FunctionStatusWhere it lives
expand_intonation_aug_coefficient Ran
ran on a synthesized input
lukewys/chamber-ensemble-generator/synth_params_augmentation.py
code served (permissive licence) · get_code("461bfc4d5af1063e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 3, dtype family float, shape (1, 14, 1) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
import pandas as pd

def expand_intonation_aug_coefficient(coefficient, conditioning_df):
    """Expand note-wise intonation augmentation coefficient to frame-wise"""
    total_length = conditioning_df.tail(1)['offset'].values[0]
    coefficient_frame_wise = np.zeros((1, total_length, 1), dtype=np.float32)
    for i, note in conditioning_df.iterrows():
        on = int(note['onset'])
        off = int(note['offset'])
        coefficient_frame_wise[:, on:off + 1, :] = coefficient[i]
    return coefficient_frame_wise


# Create a small conditioning dataframe with onset and offset columns
conditioning_df = pd.DataFrame({
    'onset': [0, 5, 10],
    'offset': [4, 9, 14]
})

# Create coefficient array with shape (num_notes, 1)
# We have 3 notes, so coefficient should have shape (3, 1)
coefficient = np.array([[0.5], [0.7], [0.3]], dtype=np.float32)

ARGS = {
    'coefficient': coefficient,
    'conditioning_df': conditioning_df
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family float, shape (1, 14, 1) — shape (1, 14, 1) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_f0 Ran
ran on a synthesized input
lukewys/chamber-ensemble-generator/data_postprocess/postprocess_utils.py
code served (permissive licence) · get_code("06c586ada97c93a0")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# The function expects synthesis_parameters to be a dict where:
# - Keys are arbitrary identifiers
# - Values are dicts with at least a 'f0_hz' key

synthesis_parameters = {
    'note_1': {'f0_hz': 440.0, 'amplitude': 0.5, 'duration': 1.0},
    'note_2': {'f0_hz': 880.0, 'amplitude': 0.3, 'duration': 0.5},
    'note_3': {'f0_hz': 220.0, 'amplitude': 0.7, 'duration': 2.0},
}

ARGS = {
    'synthesis_parameters': synthesis_parameters
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_instrument_by_part Ran
ran on a synthesized input
lukewys/chamber-ensemble-generator/utils/instrument_utils.py
code served (permissive licence) · get_code("a223123373a00d22")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

np.random.seed(0)

ARGS = {
    'ensemble': 'string',
    'part': 'Soprano'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
json_load Ran
ran; our contract draft was wrong, not the code
lukewys/chamber-ensemble-generator/utils/file_utils.py
code served (permissive licence) · get_code("29dd2f76070619d6")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import json
import tempfile
import os

# Create a temporary JSON file with sample data
temp_dir = tempfile.mkdtemp()
json_path = os.path.join(temp_dir, "test_data.json")

sample_data = {
    "key1": "value1",
    "key2": [1, 2, 3],
    "key3": {"nested": "object"}
}

with open(json_path, 'w') as f:
    json.dump(sample_data, f)

ARGS = {
    "json_path": json_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
note_expression_edit Ran
ran on a synthesized input
lukewys/chamber-ensemble-generator/expression_augmentation.py
code served (permissive licence) · get_code("25b8a3f051bc0177")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
import pandas as pd

class Config:
    def __init__(self):
        self.vibrato_range = [0.0, 1.0]
        self.volume_range = [0.1, 0.9]
        self.volume_fluctuation_range = [0.05, 0.5]
        self.volume_peak_position_range = [0.2, 0.8]
        self.attack_level_range = [0.0, 0.7]

# Create sample conditioning dataframes with required columns
def create_sample_conditioning_df():
    data = {
        'pitch': np.array([60, 62, 64, 0]),
        'vibrato': np.array([0.1, 0.2, 0.15, 0.0]),
        'volume': np.array([0.5, 0.6, 0.55, 0.0]),
        'vol_fluc': np.array([0.2, 0.25, 0.22, 0.0]),
        'vol_peak_pos': np.array([0.3, 0.4, 0.35, 0.0]),
        'attack': np.array([0.1, 0.15, 0.12, 0.0])
    }
    df = pd.DataFrame(data)
    return df

conditioning_df_all = [create_sample_conditioning_df(), create_sample_conditioning_df()]
config = Config()

ARGS = {
    'conditioning_df_all': conditioning_df_all,
    'config': config
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
pickle_load Ran
ran on a synthesized input
lukewys/chamber-ensemble-generator/utils/file_utils.py
code served (permissive licence) · get_code("e8b1c08cfe12d960")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import pickle
import tempfile
import os

# Create a temporary pickle file for testing
temp_dir = tempfile.mkdtemp()
temp_path = os.path.join(temp_dir, "test.pkl")

# Create some sample data and pickle it
sample_data = {
    "key1": "value1",
    "key2": [1, 2, 3],
    "key3": {"nested": "dict"}
}

with open(temp_path, 'wb') as f:
    pickle.dump(sample_data, f)

ARGS = {
    "path": temp_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
split_piece_list Ran
ran on a synthesized input
lukewys/chamber-ensemble-generator/data_postprocess/postprocess_and_unchunk_cocochorales.py
code served (permissive licence) · get_code("5c6b55d5270959cb")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import os
import tempfile

# Create a temporary directory structure to simulate file paths
temp_dir = tempfile.mkdtemp()

# Create some dummy midi files (just empty files for testing)
piece_files = [
    os.path.join(temp_dir, "piece_001.mid"),
    os.path.join(temp_dir, "piece_002.mid"),
    os.path.join(temp_dir, "piece_003.mid"),
    os.path.join(temp_dir, "piece_004.mid"),
]

for f in piece_files:
    open(f, 'a').close()

# Create piece_list with full paths
piece_list = piece_files

# Create split_json that references the basenames (without .mid extension)
split_json = {
    "train": ["piece_001.mid", "piece_002.mid"],
    "val": ["piece_003.mid"],
    "test": ["piece_004.mid"],
}

ARGS = {
    "piece_list": piece_list,
    "split_json": split_json,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
yaml_load Ran
ran on a synthesized input
lukewys/chamber-ensemble-generator/utils/file_utils.py
code served (permissive licence) · get_code("a40695b615185181")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tempfile
import os
import yaml

# Create a temporary YAML file with test data
temp_dir = tempfile.mkdtemp()
yaml_path = os.path.join(temp_dir, "test.yaml")

test_data = {
    "key1": "value1",
    "key2": 42,
    "key3": [1, 2, 3],
    "nested": {
        "inner_key": "inner_value"
    }
}

with open(yaml_path, 'w') as f:
    yaml.dump(test_data, f)

ARGS = {
    "yaml_path": yaml_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
yaml — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
assign_expressive_performance Not yet run
unverified (harvested, no recorded run)
lukewys/chamber-ensemble-generator/midi_augmentation.py
code served (permissive licence) · get_code("d7fa9419744b6cd8")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 8 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
json_load 2 other papers
2 later
2401.01044 (2024-01)
Auffusion: Leveraging the Power of Diffusion and Large Language Models for Text-to-Audio Generation
2409.12466 (2024-09)
AudioEditor: A Training-Free Diffusion-Based Audio Editing Framework

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Data is the lifeblood of modern machine learning systems, including for those in Music Information Retrieval (MIR). However, MIR has long been mired by small datasets and unreliable labels. In this work, we propose to break this bottleneck using generative modeling. By pipelining a generative model of notes (Coconet trained on Bach Chorales) with a structured synthesis model of chamber ensembles (MIDI-DDSP trained on URMP), we demonstrate a system capable of producing unlimited amounts of realistic chorale music with rich annotations including mixes, stems, MIDI, note-level performance attributes (staccato, vibrato, etc.), and even fine-grained synthesis parameters (pitch, amplitude, etc.). We call this system the Chamber Ensemble Generator (CEG), and use it to generate a large dataset of chorales from four different chamber ensembles (CocoChorales). We demonstrate that data generated using our approach improves state-of-the-art models for music transcription and source separation, and we release both the system and the dataset as an open-source foundation for future work in the MIR community.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2209.14458")
get_code_for_paper("2209.14458")
have("2209.14458")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2209.14458.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2209.14458.svg)](https://syntology.ai/paper/2209.14458)

Connect an agent — have() is free.