Audio

Log and view audio data in Pluto

Logging

Audio files can be logged via a file path or a NumPy array containing audio data shaped as frames × channels.

To log audio, instantiate the pluto.Audio class:

audio = pluto.Audio(
    data=Union[str, np.ndarray],
    rate=int | None = 48000,
    caption=str | None = None,
)
pluto.log({"audio/0": audio}, step=step)

Parameter	Type	Description
data	Union[str, np.ndarray]	The audio data to log. Can be a path to an audio file or a NumPy array.
rate	int	The sample rate of the audio data. Defaults to 48000.
caption	str	A caption for the audio.

Examples

Logging from File Paths

import httpx
r = httpx.get(
    "https://actions.google.com/sounds/v1/alarms/digital_watch_alarm_long.ogg"
)
with open(f"test.ogg", "wb") as f:
    f.write(r.content)

pluto.log({"audio": pluto.Audio(data="test.ogg")}, step=step)

Logging from NumPy Arrays

data = np.array([[1, 1, 1], [1, 1, 1]], dtype=np.float32)
pluto.log({"audio": pluto.Audio(data=data)}, step=step)

Viewing

Logged audio files appear as player widgets with playback controls, volume adjustment, and per-file Analyze and Download buttons. When comparing multiple runs, each audio card shows the run name with its assigned color.

Comparing multi-sample audio logs across runs.

Adding to a dashboard

Audio players show up automatically in the Compare view. To place them on a custom dashboard, open Add Widget, choose the Files tab, and search for your audio log name.

Playback Controls

Each audio player includes:

• Play / Pause with a progress slider you can drag to seek
• Skip forward / back buttons (5-second jumps)
• Volume slider with mute toggle
• Download — saves the audio file locally
• Analyze — opens the audio analysis dialog (see below)

Step Navigation

If you log audio at multiple training steps, use the step slider below the players to browse through different steps. This is useful for tracking how generated audio (e.g., text-to-speech) improves over the course of training.

When multiple audio groups are displayed in the same section, their step sliders can be linked so that changing the step on one group changes all of them simultaneously. Click the lock icon on the step navigator to toggle sync on or off.

Multi-Sample Logging

Log several clips at the same step under one log name by passing a list. Every clip renders inside the same card with a per-cell ◀ i / N ▶ nav row below the player:

pluto.log({
    "samples": [pluto.Audio(a), pluto.Audio(b)],
}, step=10)

Sync Sample Indices

A Sync Sample Indices selector in the media settings popover controls how the sample index is shared:

• Off — each run steps through its samples independently.
• Across runs — every run in the same widget shares the sample index.
• Across widgets (default) — every run on every media widget (image, video, and audio) advances the sample index in tandem.

The selector is always available, so you can opt a single widget out of cross-widget sync. Where a card has fewer samples than the shared index, it clamps to its own last sample, and single-sample cards show no nav row.

Pinning Runs at a Step

The global step slider forces every run to be viewed at the same step, which makes it hard to compare runs at their respective best checkpoints — different runs converge at different steps. Pinning lets you freeze individual runs at a specific step while the global slider keeps driving everything else.

Each pinned card shows a colored badge with the pinned step. Click the pin button on an audio card to choose the pin scope:

Symbol	Scope	Effect
◇	Pin in this panel	Freezes only this card’s run inside this audio widget. Other widgets for the same run still follow the global stepper.
◈	Pin across all panels	Freezes this run at the chosen step in every audio widget on the page that shows it.
★	Pin to best step	Pinned by a column-header action (see Best-Step Pinning below), computed per run from a metric’s argmin/argmax.

Unpinning

Cross-panel and best-step pins offer two unpin options: Unpin this audio removes the pin from this widget only, while Unpin across all panels removes it everywhere. Local single-panel pins keep the simpler one-click unpin.

Clear All Pins

Click Clear All Pins to remove every pin (local, cross-panel, and best-step) at once for the current page.

Best-Step Pinning

For metric-driven exploration (e.g. “play each run at its best loss”), pin every run on the page to its own argmin/argmax step in one click. Open a metric column’s header menu in the runs table, then choose Find best step → one of:

Action	What it pins to
Pin steppers at min value	Each run pinned at the step where the metric is lowest
Pin steppers at max value	Each run pinned at the step where the metric is highest
Pin steppers at min (with media)	Lowest metric value among steps that also have a clip for that run
Pin steppers at max (with media)	Highest metric value among steps that also have a clip for that run

The (with media) variants (grouped under Pin to nearest media step) are useful when your metric and audio cadences don’t perfectly overlap — without them, the chosen step might land on a step that has no clip, so the card falls back to the No audio at step N placeholder.

Tolerance Window

Even with the (with media) variants, your metric and audio cadences may not align exactly. The best-step picker handles this with a nearest-snap with tolerance: each metric step snaps to its closest clip step within a step-distance limit K (default K = 20).

You can edit K directly in the Find best step submenu — type a value, press Enter or click any of the four pin buttons.

Widen the window if pins come up empty. In the tol-demo-gap run below, the metric is logged at steps 0/50/100/150/200 and clips only at 25/75/125/175, so every metric step sits exactly 25 steps from the nearest clip. At the default K = 20 nothing is within tolerance and the run stays unpinned (plain pin icon, no step badge); raise K to 25 and the min-loss step (100) snaps to audio step 125 — the provenance even notes the tie with step 75, broken toward the later step.

A wider window can also find a better step. When the metric keeps improving past the nearest media step, a tight window snaps to a worse value. In tol-demo-sparse below, K = 10 only reaches metric step 40 (loss 0.85), while K = 25 reaches metric step 55 — the true minimum (loss 0.1). The pinned clip is the same in both (step 30); only the provenance popover reveals which metric step drove the pick.

Pin Provenance

Every best-step pinned card shows an info icon (ⓘ) next to the run name. Hover for a popover that explains why this step was chosen:

• A headline with the metric and extreme that drove the pick (e.g. Pinned at min train/loss = 0.006667)
• The metric step that drove the pick
• How far the chosen clip is from that metric step (or that it matches exactly)

Audio Analysis

Click Analyze on any audio player to open a dialog with three tabs:

• Spectrum — Real-time frequency spectrum visualization that animates during playback
• Waveform — Time-domain waveform drawn from the audio buffer
• Statistics — Peak amplitude, RMS level, duration, sample rate, number of channels, and dynamic range