Description #

Speech endpoint detection module – an efficient audio segmentation solution based on FSMN VAD

Key features #

• Speech endpoint detection: Automatically identifies the start and end times of speech segments in audio, enabling precise segmentation between speech and silence
• Batch processing: Supports efficient endpoint detection for large volumes of audio data
• Multi-format input: Compatible with various audio input methods, including raw binary, Base64 encoding, and TOS/HTTP links
• GPU acceleration: Supports high-performance inference in GPU environments

Recommended practices #

• It is recommended to input WAV format audio with a 16k sample rate and mono channel to improve detection accuracy
• For long audio files, segment processing is recommended; the duration for a single processing should not exceed 1 hour
• Applicable to scenarios such as voice activity detection and speech segmentation

Supported models #

• iic/speech_fsmn_vad_zh-cn-16k-common-pytorch (General Chinese FSMN VAD)

Output description #

• Each audio outputs a two-dimensional floating-point list representing the start and end timestamps of all speech segments (unit: seconds), for example: [[0.0, 4.34], [5.50, 7.12]]
• If processing fails, returns None

Operator parameters #

Input #

Output #

A column containing speech endpoint timestamps; each element is a two-dimensional floating-point list representing the start and end timestamps of each speech segment in the audio (unit: seconds),
for example: [[0.0, 4.34], [5.50, 7.12]]. If processing fails, returns None.

Parameters #

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

Examples #

The following code demonstrates how to use daft to run the operator and detect speech endpoints in audio.

from __future__ import annotations

import logging
import os

import ray

import daft
from daft import col
from daft.las.functions.audio.audio_vad_fsmn import AudioVadFsmn
from daft.las.functions.udf import las_udf

if __name__ == "__main__":

    if os.getenv("DAFT_RUNNER", "ray") == "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.%s".format(),
            )
            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)

        import ray

        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)

    tos_dir_url = os.getenv("TOS_DIR_URL", "las-cn-beijing-public-online.tos-cn-beijing.volces.com")
    samples = {
        "audio_path": [
            f"https://{tos_dir_url}/public/shared_audio_dataset/.wav"
        ]
    }

    model_path = os.getenv("MODEL_PATH", "/opt/las/models")
    model_name = "iic/speech_fsmn_vad_zh-cn-16k-common-pytorch"
    audio_src_type = "audio_url"
    batch_size_s = 3600
    model_revision = "v2.0.4"
    rank = 0

    df = daft.from_pydict(samples)
    df = df.with_column(
        "audio_vad_result",
        las_udf(
            AudioVadFsmn,
            construct_args={
                "audio_src_type": audio_src_type,
                "model_path": model_path,
                "model_name": model_name,
                "model_revision": model_revision,
                "batch_size_s": batch_size_s,
                "rank": rank,
            },
            num_gpus=1,
            batch_size=1,
            concurrency=1,
        )(col("audio_path")),
    )
    df.show()

    # ╭────────────────────────────────┬─────────────────────╮
    # │ audio_path                     ┆ audio_vad_result    │
    # │ ---                            ┆ ---                 │
    # │ Utf8                           ┆ List[List[Float32]] │
    # ╞════════════════════════════════╪═════════════════════╡
    # │ tos://las-cn-beijing-public-o… ┆ [[0.51, 2.8]]       │
    # ╰────────────────────────────────┴─────────────────────╯