Description #

Audio metadata extraction operator

Function #

1. Supported audio file formats
2. Extract complete audio metadata information, including:

• Basic information: duration, format, bit rate
• Audio information: sample rate, number of channels, encoder

3. Use the ffprobe tool to extract metadata
4. Supports local files, remote files (TOS/S3), HTTP/HTTPS links

Return #

Structured array containing the following fields:

• duration (float): audio duration (seconds)
• format_name (string): format name
• bit_rate (int64): bit rate
• audio_codec (string): audio encoder name
• audio_sample_rate (int32): audio sample rate
• audio_channels (int32): number of audio channels

Operator parameters #

Input #

Output #

Structured array containing metadata

Parameters #

If a parameter does not have a default value, it is required

Examples #

The following code demonstrates how to use Daft (for distributed environments) to run the operator and extract metadata information from audio files.

from __future__ import annotations

import os

import daft
from daft import col
from daft.las.functions.audio import AudioMetadataExtract
from daft.las.functions.udf import las_udf

if __name__ == "__main__":
    TOS_TEST_DIR_URL = os.getenv("TOS_TEST_DIR_URL", "las-cn-beijing-public-online.tos-cn-beijing.volces.com")

    if os.getenv("DAFT_RUNNER", "native") == "ray":
        import logging

        import ray

        def configure_logging():
            logging.basicConfig(
                level=logging.INFO,
                format="%(asctime)s - %(name)s - %(levelname)s - %(message)s",
                datefmt="%Y-%m-%d %H:%M:%S",
            )
            logging.getLogger("tracing.span").setLevel(logging.WARNING)
            logging.getLogger("daft_io.stats").setLevel(logging.WARNING)
            logging.getLogger("DaftStatisticsManager").setLevel(logging.WARNING)
            logging.getLogger("DaftFlotillaScheduler").setLevel(logging.WARNING)
            logging.getLogger("DaftFlotillaDispatcher").setLevel(logging.WARNING)

        ray.init(dashboard_host="0.0.0.0", runtime_env={"worker_process_setup_hook": configure_logging})
        daft.set_runner_ray()

    daft.set_execution_config(actor_udf_ready_timeout=600)
    daft.set_execution_config(min_cpu_per_task=0)

    samples = {
        "input_path": [
            f"https://{TOS_TEST_DIR_URL}/public/archive/audio_metadata_extract/sample.wav",
        ],
    }
    ds = daft.from_pydict(samples)

    # Using Daft to extract audio metadata
    constructor_kwargs = {
        "timeout": 600,
    }

    ds = ds.with_column(
        "metadata",
        las_udf(
            AudioMetadataExtract,
            construct_args=constructor_kwargs,
            num_cpus=1,
            concurrency=1,
            batch_size=1,
        )(col("input_path")),
    )

    ds.show()
    # ╭────────────────────────────────┬──────────────────────────────────────────────────────────────────────────────╮
    # │ input_path                     ┆ metadata                                                                     │
    # │ ---                            ┆ ---                                                                          │
    # │ String                         ┆ Struct[duration: Float64, format_name: String, bit_rate: Int64, audio_codec: │
    # │                                ┆ String, audio_sample_rate: Int32, audio_channels: Int32]                     │
    # ╞════════════════════════════════╪══════════════════════════════════════════════════════════════════════════════╡
    # │ https://las-public-data-qa.to… ┆ {duration: 49.71102,                                                         │
    # │                                ┆ format_n…                                                                    │
    # ╰────────────────────────────────┴──────────────────────────────────────────────────────────────────────────────╯