audio_processing

Releases

1. ▸▾0.2.06 Aug 2026

   Add AecDelayEstimator: envelope cross-correlation delay estimation between a loopback reference and a microphone capture, with recency-weighted median

   • Add AecDelayEstimator: envelope cross-correlation delay estimation between a loopback reference and a microphone capture, with recency-weighted median smoothing and two lock tiers. Amplitude thresholds are float32-native; see AecDelayEstimator.defaultMinNearStd.
   • Add AecBlockAccumulator plus floatToPcm16/pcm16ToFloat for chopping float32 captures into the exact 10 ms int16 blocks an echo canceller requires.
   • Add mixTracksToMono and peakBuckets for offline mixdown and waveform rendering of completed recordings of differing lengths, which the streaming AudioMixer cannot express.
   • Adapted from Control Center (MIT © 2026 Samuel Alev); see NOTICE.
   • Add AudioRing: a fixed-capacity float32 ring addressed by a monotonic absolute write index, so a position reported by a VAD — counted from the start of the stream — indexes the audio directly no matter how often the buffer has wrapped. Ranges older than the capacity are clipped rather than wrapped, because returning samples from the wrong moment is indistinguishable from success. Defaults to 160,000 samples, ten seconds at 16 kHz.
   • Add AdaptiveGain and AdaptiveGainProcessor: slow automatic gain control that amplifies quiet capture toward a 0.3 target peak (30x ceiling, 0.995 per-chunk running-peak decay, 1.05 deadband) and never attenuates. Speech models trained on normalized audio collapse on the 0.05-peak recordings real microphones produce; the slow decay is what keeps the noise floor from being pumped up between utterances. The processor keeps one running peak per stream and drops it on a discontinuity.
   • Add segmented WAV recording: SegmentedWavRecorder rotates bounded WAV segments while durably mirroring a JSON sidecar (SegmentedWavRecordingManifest) with pause/resume/source-change markers, bounded queue policies (wait, dropNewest, failRecorder), and a stable failed state on segment IO errors. Storage is injected through the new SegmentedWavStorage/SegmentedWavStorageFile ports; LocalSegmentedWavStorage is the dart:io adapter.
   • Add crash recovery for segmented recordings: recoverSegmentedWavRecording repairs every segment named by the durable sidecar after an interrupted process, and repairWavStorageFile repairs one canonical WAV held in any SegmentedWavStorageFile. repairWavFileHeader covers the path-based dart:io case.
   • Add WavFileAudioSource: an AudioSource over a canonical WAV file for offline pipelines.
   • Keep the portable/dart:io barrel split intact for the new APIs: audio_processing.dart exports only portable code (recorder, manifest, storage ports, storage-based recovery), while audio_processing_io.dart adds LocalSegmentedWavStorage, WavFileAudioSource, WavFileAudioSink, and repairWavFileHeader. SegmentedWavRecorder.create and recoverSegmentedWavRecording take required storage instead of a directoryPath convenience, so portable code never constructs dart:io storage implicitly.

• audio_corepub.devProvider-neutral PCM audio types and asynchronous session contracts.
• audio_flutterpub.devProvider-neutral Flutter microphone, system-audio, and PCM playback.
• audio_flutter_darwinpub.devDarwin implementation of audio_flutter capture and PCM playback.
• audio_flutter_linuxpub.devLinux implementation of audio_flutter capture and PCM playback over PulseAudio and PipeWire.
• audio_flutter_platform_interfacepub.devTyped platform contract for audio_flutter capture and playback.