A video export can sound fine on one playback system and still contain an unexpectedly loud passage, a long silent gap or a stereo problem. Before a take, the question is different: is this microphone setup usable here? After an edit, it becomes: did the new export actually improve?
SignalMetric 2.0 brings these questions into one local audio-checking workflow on iOS and Android. Both release candidates use 2.0.0, build 2, with matching core tasks, measurement references and report actions.
These are the feature notes for the 2.0 release candidate, not an announcement that both store releases are available. As of September 15, 2026, the release record does not confirm a public 2.0 rollout. Android is coming soon; check the official store listing for iOS availability.
An easier starting point
The home screen now gives professional instruments a prominent, full-width entry, followed by dedicated tasks for environment noise, checking an export, preparing a recording and comparing edits.
The title and settings share one row, language selection lives in settings, and recent checks are collapsed by default. The layout reduces scrolling on everyday phone sizes while retaining adaptive layouts for narrow screens, tablets, landscape and larger system text.
For a first check, an explicitly labeled, eight-second synthetic audio example runs through the actual file analyzer. It needs neither a personal file nor microphone permission, and it does not start playback automatically. The example is a way to inspect a real result before bringing in your own media.
Check the whole export before publishing
Import a supported audio file, or the audio track of a supported video, and analyze the complete file. A batch can contain up to ten items, each with its own progress and error state. Cancelling pending work retains the results already completed.
The report brings together:
- Integrated loudness and its difference from the selected reference.
- Sample peak and a 4x True Peak estimate.
- Near-full-scale passages and silent intervals with time positions.
- Stereo balance, correlation and estimated loss when folded to mono.
- Suggested gain and the estimated peak after that gain change.
Select an event or an interval in the envelope to locate what needs another listen. The purpose is to replace an unexplained number with evidence you can take back to the edit.
The video reference uses -14 LUFS / -1 dBTP; the lesson and podcast reference uses -16 LUFS / -1 dBTP. EBU R128 and ATSC A/85 references remain selectable. These are comparison targets, not universal platform requirements or delivery certification. Suggested gain is advice: SignalMetric does not normalize, repair or overwrite your source audio.
Check the setup before recording
Recording preparation separates a five-second environment sample from a ten-second voice sample. That makes the comparison more useful than treating a quiet room and a spoken phrase as one measurement.
The check considers the observed floor, voice level and available peak margin. It also checks that the samples belong to the same input setup and are fresh enough to use. A source change or dropped analysis windows can invalidate the check instead of quietly producing a reassuring result.
Use it to evaluate a mic position or recording location before the main take. It remains a setup check, not a guarantee that later speech, traffic or handling noise will stay within the sampled conditions.
Keep the before and after
Save a measurement with a title and notes, then compare it with a later export. A/B comparison puts the measurements and spectral evidence alongside one another so you can inspect what changed after adjusting gain, editing pauses or revising a mix.
Swipe horizontally to explore the diagram
This diagram could not be rendered. Its source is still available below.
View diagram source
flowchart TD
A["Check the original export"] --> B["Save the measurement"]
B --> C["Edit in your audio or video editor"]
C --> D["Check the revised export"]
D --> E["Compare A and B"]
E --> F["Export a PDF, CSV or JSON report"]Changing the reference for a saved result creates a new draft rather than replacing the original evidence. Revision checks reject stale saves, and deletion requires explicit confirmation. PDF, CSV and JSON exports let you keep or share the measurement separately from the media.
These exports do not introduce cloud sync, and the JSON files are not a cross-platform archive-import format.
A separate environment-noise tool
Environment noise is now its own task. Use it to follow changes in a room, recording corner or parked car without creating an audio recording.
The display includes a live A-weighted reading, FAST response, an energy-equivalent level, minimum and maximum readings, and a history trace. The unit matters:
- Before calibration: the reading is relative dBFS(A), referenced to the digital input's full scale. It is not the room's absolute sound-pressure level.
- After calibration: comparison with a trusted sound-level meter enables estimated dBA, including an estimated LAeq reading.
- When the input route changes: the calibration no longer applies to the new route and must be re-established.
Phone microphones, gain processing and external interfaces differ. A calibration offset does not turn a phone into a certified sound-level meter or validate its response across every frequency and level. This tool is not intended for hearing-safety decisions, occupational compliance or legal noise evidence.
The noise session analyzes the input without saving an audio file.
Instruments that are easier to read
The professional workspace keeps its four views: Monitor, Spectrum, Timeline and Scope. The visual update adds segmented channel-level meters with peak hold, clearer spectrum grids and peak markers, and a graduated analog-style meter.
Demo, microphone and local-file sources remain available, alongside explicit recording, reference capture and display freeze. Six skins let the same workspace fit different viewing preferences without changing the measurement meaning.
Both apps support English, Simplified Chinese, Traditional Chinese, Japanese and Korean. This blog retains the site's four article languages: English, Simplified Chinese, Japanese and Korean.
What alignment means on Android
Android now follows the same core workflow as iOS: file and batch checks, recording preparation, saved measurements, comparison, exports, environment noise and the professional instruments.
This iteration also improves Android-specific behavior:
- Microphone recovery when returning to the instrument workspace after backgrounding.
- Loop playback enabled by default for local audio, with protection against outdated playback callbacks restarting an old source.
- Playback and loop controls in one compact row, without separate skip buttons.
- Smaller default type and adaptive control arrangements for narrow screens.
- Recoverable recording data if AAC encoding fails, and stricter handling of stale report saves.
Alignment does not mean identical system dialogs or identical microphone hardware. Document pickers, permissions, device audio processing and supported codecs remain platform-dependent. Android external imports have a 500 MiB limit.
The same PCM fixtures passed 58 Swift/Kotlin comparisons across selected mono and stereo sample rates, checking loudness, peaks, channel relationships and event counts. That is evidence for those fixtures, not proof that every codec, phone or external microphone produces identical results.
Local by design, with separate store availability
The product model is unchanged: one-time purchase, no ads and no subscriptions. Audio analysis stays on the device, no account is required, and audio is not uploaded for processing. Recording remains an explicit action; submitting support feedback or opening a website is a separate network action.
iOS and Android purchases are platform-specific. There is no shared purchase entitlement or automatic cross-platform library sync.
The 2.0 candidates align the work you can do, while store distribution and remaining physical-device checks are separate release steps. Visit the SignalMetric website for availability, check the iOS App Store listing, or explore the SignalMetric product guide.
For background on the measurements, What an Audio Meter Is Actually Telling You explains why level, loudness, frequency and time answer different questions. The 2.0 workflow puts those distinctions to work before publishing, before recording and after editing.