CHANGELOG.md

Version History

Version 1.1.0 (2024-12-11)

• Increase MSRV to 1.76
• Greatly improved the ConvolverNode performance for large impulse responses
• Fix AudioBufferSourceNode reverse playback
• Implement sub-sample AudioBufferSourceNode stitching

Version 1.0.1 (2024-09-18)

• AudioBufferSourceNode: handle reverse playback and looping
• AudioBufferSourceNode: handle sub-sample stop/duration times
• AudioBufferSourceNode: fix handling of out of bound offsets

Version 1.0.0 (2024-09-10)

• Breaking: MediaRecorder constructor now requires options (e.g. mime type)
• Added MediaRecorder::is_type_supported to check if recording is supported for this mime type
• Fixed the up/down-sampling of the WaveShaperNode
• Fixed an infinite loop in AudioBufferSourceNode when using out-of-bound loop start/end values
• Fixed AudioBufferSourceNode crashing or restarting on certain loop value changes

Version 0.45.2 (2024-06-12)

• Fix excessive buffer size (latency) on Android

Version 0.45.1 (2024-06-04)

• Fixed a crash for AudioWorkletNode with zero outputs
• Increase MSRV to 1.71

Version 0.45.0 (2024-05-06)

• Changed: AudioContext::render_capacity now returns owned AudioRenderCapacity instead of ref
• Added AudioParamValues.keys iterator

Version 0.44.0 (2024-04-22)

• Fix regression: ensure all async fns return a Send + Sync Future object

Version 1.0.0-rc.6 (2024-04-20)

• Breaking: the ScriptProcessorNode::set_onaudioprocess no longer takes a ref to the event
• Fixed a panic for multi-channel output with the ScriptProcessorNode
• Fixed an issue when calling suspend/close on an already suspended context

Version 0.43.0 / 1.0.0-rc.5 (2024-04-15)

• Changed: renamed AudioNode::connect_at to connect_from_output_to_input
• Changed: renamed AudioNode::disconnect_from to disconnect_dest
• Added AudioNode::disconnect_output and disconnect_dest_from_output(to_input)
• Fix: AudioNode::disconnect() would clear all connections instead of only outgoing ones
• Fix: A panic is now thrown when attempting to disconnect an unconnected node

Version 1.0.0-rc.4 (2024-04-13)

• Changed: enforce minimum 2k and maximum 384k Hertz sample rate for audio contexts
• Added the ScriptProcessorNode
• Fix: all events are now dispatched for OfflineAudioContext
• Fix: BiquadFilter/IIRFilterNode freq response return NaN for invalid frequencies
• Fix: do not allow length 0 for OfflineAudioContext and AudioBuffer

Version 1.0.0-rc.3 (2024-04-02)

• Breaking: ChannelConfigOptions is renamed to AudioNodeOptions
• Fix WaveShaperNode: do not emit silence for non-centric curve
• Fix DynamicsCompressorNode: handle channel constraints
• Fixed error messages and constraints for DestinationNode and ChannelSplitter/MergerNode

Version 1.0.0-rc.2 (2024-03-07)

• Changed: the AudioContext will continue playing if it is dropped (use close() instead)
• Apply AudioParamDescriptor validations
• Apply limits of compound AudioParams
• Fix AudioParam rounding errors
• Fix k-rate AudioParams not always delivering constant value
• Fixed a panic while rendering AudioParams with dynamic lifetimes
• Disallow constructing an AudioBuffer with length zero

Version 1.0.0-rc.1 (2024-02-24)

• Breaking: the AudioProcessor trait is now superseded by AudioWorkletProcessor
• Breaking: the RenderScope struct is renamed to AudioWorkletGlobalScope
• Added message port functionality to AudioWorkletNode and AudioWorkletProcessor
• Added OfflineAudioContext event handlers
• Added validation of AudioScheduledSourceNode start/stop actions
• Fixed a bug with min/max decibels in the AnalyserNode
• Fixed panic in AudioBufferSourceNode when duration exceeds buffer length

Version 0.42.1 (2024-02-12)

• Add Debug implementation for all public facing items
• Fix: return proper channelCountMode for StereoPannerNode

Version 0.42.0 (2024-02-05)

• Added MediaTrackConstraints.channelCount for multi-channel microphone input
• Fix: needless upmixing of AudioNode input channels for max and clampedMax interpretation
• Fix: AudioParam value not returned correctly before rendering has started
• Fix: eagerly panic for invalid AudioNode settings in constructors

Version 0.41.1 (2024-01-11)

• Unified panic messages to format <JsErrorType> - <Actual message>
• Unified README.md and crate level docs
• Fix: minimize floating point errors in AudioBufferSourceNode currentTime calculation
• Fix: AudioParam automationRate does not instantly update on the control thread
• Fix: AudioParam better handling of non finite values in arguments
• Fix: Panic at Analyser initialization for some combinations of min/maxDecibel

Version 0.41.0 (2024-01-04)

• Fix BaseAudioContext onChangeState to signal device readiness
• Fix OfflineAudioContext::state is not updated when calling suspend and resume
• Fix AudioContext resume_sync, suspend_sync, close_sync now block until completion
• Added async versions of AudioContext suspend/resume/close

Version 0.40.0 (2024-01-02)

• Assert AudioScheduledSourceNode start and stop time values are valid
• Added BaseAudioContext clearOnstatechange and fix some event type names

Version 0.39.1 (2023-12-28)

• Fix AudioWorkletNode custom output channel counts

Version 0.39.0 (2023-12-27)

• Added async OfflineAudioContext methods: startRendering, suspend and resume
• Added sync OfflineAudioContext methods: suspend
• Added BaseAudioContext::onStateChange

Version 0.38.0 (2023-12-03)

• Breaking: OfflineAudioContext::start_rendering_sync takes &mut instead of self
• Fix panic in AudioBufferSource when supplying buffer in options
• Fix panic in Oscillator when supplying periodic wave in options
• Fix panic in AudioContext for really low sample rates
• Support up/down-mixing of all speaker layouts with 6 channels or less
• Disallow channel merger/splitter configurations with more than 32 channels

Version 0.37.0 (2023-11-25)

• Added the AudioWorkletNode and AudioWorkletProcessor interfaces
• Added stereo-to-stereo panning for PannerNode in EqualPower mode
• Aggressively release resources of 'orphaned' audio nodes
• Tuned the volume of stereo-to-stereo panning for PannerNode in HRTF mode
• Added context.run_diagnostics for debugging purposes
• Fixed a panic that could occur during audio graph booting (flawed is_active check)
• Fix: disconnect would disassociate an AudioNode from its AudioParams
• Fix: properly clamp PannerNode rollOffFactor according to the distance model

Version 0.36.1 (2023-11-08)

• Fix compilation on targets that are not x86/aarch64
• Added the ability to run benchmarks locally with criterion

Version 0.36.0 (2023-10-20)

• Disable floating point denormals in audio processing via processor flags
• Do not spawn a garbage collector thread for an OfflineAudioContext
• AudioParam now derives Clone

Version 0.35.0 (2023-10-18)

• Fix: panic when creating and dropping PannerNodes
• Improved performance of setting panning mode to HRTF

Version 0.34.0 (2023-10-12)

• Breaking: many AudioNode setter methods now require &mut self instead of &self
• Fix: all audio node settings are now applied in order in the audio graph
• Fix: render thread would crash when a PannerNode is removed
• Fixed and improved device ids for audio input/output devices
• Added current playhead position for AudioBufferSourceNode

Version 0.33.0 (2023-07-27)

• Fix: clamp to number of output channels to 32 even if the hardware supports more to prevent runtime panics
• Fix: prevent a render thread panic if the audio graph is not fully initialized yet
• Fix: rename AudioDestination max_channels_count to max_channel_count
• Fix: AudioBufferSourceNode is now Send + Sync
• Change AudioProcessor::onmessage signature to prevent deallocations
• Add garbage collector thread as sidecar of the render thread to handle some deallocations

Version 0.32.0 (2023-07-16)

• Fix for some audio node settings being applied out of order
• Fix setting the ChannelCountMode for the OfflineAudioContext destination node
• Fix and extend AAC/M4A/ALAC decoding
• Fix broken MediaElementSourceNode for multi-channel output
• Updated Minimum Supported Rust Version (MSRV) to 1.70

Version 0.31.0 (2023-06-25)

• Avoid allocations in Waveshaper node for real-time safety
• Improvements for the JACK audio host
• Document ALSA limitations for low latencies, and add fallback to examples
• Improve real-time safety of the render thread by using bounded channels
• Improve our usage of atomics
• Update to Rust edition 2021

Version 0.30.0 (2023-06-07)

• Implement MediaRecorder API
• AudioContext now uses default stereo channels
• Fix issue with AudioBufferSourceNode playback rate and detune
• Rename enumerate_devices to enumerate_devices_sync
• Don't panic on unavailable input/output device selection

Version 0.29.0 (2023-05-07)

• Implement part of the MediaStreams API (MediaStream, MediaStreamTrack)
• Implement part of the MediaDevices API (getUserMedia, enumerateDevices)
• Changed enumerateDevices to also include input devices
• Microphone input can now specify desired deviceId
• Added MediaStreamTrackAudioSourceNode
• Fixed windows build by removing termion from dev-dependencies

Version 0.28.0 (2023-01-30)

• Improved AnalyserNode (performance and correctness)
• Fixed microphone input on Raspberry Pi
• Fixed overriding channel config for PannerNode

Version 0.27.0 (2023-01-17)

• Head Related Transfer Function (HRTF) panning mode
• Implemented event handlers
• Prepare API for variable render quantum sizes

Version 0.26.0 (2022-11-13)

• Added the "none" sinkId, render audio graph without emitting to speakers
• Fix ConvolverNode.set_normalize to take &self, not &mut self

Version 0.25.0 (2022-11-06)

• Added AudioRenderCapacity functionality
• Added sinkId functionality to AudioContext (specify audio output device)
• Renamed ChannelConfigOptions.{mode -> count_mode}

Version 0.24.0 (2022-09-10)

• Added ConvolverNode (mono only)
• GainNode, BiquadFilterNode, AudioParam performance improvements

Version 0.23.0 (2022-08-23)

• AudioParam computed values array now contains only a single value when k-rate or no automations scheduled
• DelayNode supports sub-quantum delay
• IIRFilterNode and BiquadFilterNode can handle multi-channel inputs
• Various performance improvements

Version 0.22.0 (2022-07-29)

• Added DynamicsCompressorNode
• Added cubeb as an alternative audio backend

Version 0.21.0 (2022-07-23)

• Implemented MediaElement and MediaElementSourceNode
• Improved performance of AudioBufferSourceNode and DelayNode
• Make AudioContext and OfflineAudioContext Send and Sync
• Relaxed the 'balanced' and 'playback' latency - to run smoothly on RPi
• Fixes on clamping and value calculations of AudioParams
• Can now change automation rate of an AudioParam on the fly
• Implemented the concept of 'actively processing' for AudioNodes

Version 0.20.0 (2022-07-02)

• Change sample rate type to plain f32
• Remove namespacing of buffer, audio_param and periodic_wave
• Reduce AudioRenderQuantum public API surface
• Remove ConcreteBaseAudioContext from public API docs

Version 0.19.0 (2022-06-01)

• Added baseLatency and outputLatency attributes
• Audio processor callback now has access to AudioWorkletGlobalScope-like env
• Performance optimization for sorting large audio graphs
• Use default sample rate for output devices instead of highest to prevent insane values
• Fixed incorrect channel mixing for AudioNode input buffers

Version 0.18.0 (2022-04-12)

• Implement BaseAudioContext state
• AudioContext can now change the number of output channels while running
• Microphone input stream is now configurable
• Microphone can properly pause, resume and close
• Consistently use usize for channels, inputs, lengths

Version 0.17.0 (2022-04-03)

• Simplify AudioNode's required channel methods
• Apply all channel count/mode/interpretation constraints
• AudioContext can now update channel count while playing
• Improve AudioContext constructor
• Validate more input values for AudioParam events

Version 0.16.0 (2022-03-20)

• AudioBufferSourceNode can now resample
• Add MediaStreamAudioDestinationNode
• Performance improvement when no PannerNodes are present
• More consistent method/argument names
• Added benchmark program
• Removed MediaElement and MediaElementSourceNode for now, will reimplement

Version 0.15.0 (2022-02-10)

• Allow method chaining on AudioParams
• Some fallible methods will now panic instead of returning a Result
• Sub sample scheduling for OscillatorNode
• Implement AudioContext.close to free resources
• Rename BaseAudioContext trait and concrete type
• Fix spec deviations for node methods and constructor options and defaults
• Rename some functions to _sync to denote they do not return a Promise

Version 0.14.0 (2022-01-13)

• Implemented context.decodeAudioData
• New media decoder using symphonia crate (with MP3 support)
• WaveShaper fixes
• Spec compliance: float for sample rate, thread safe nodes, naming of nodes

Version 0.13.0 (2021-12-28)

• Support cyclic audio graphs with DelayNode acting as cycle breaker
• Greatly improved AudioBufferSourceNode (sub quantum scheduling, more controls, performance)
• Added cone gain functionality for PannerNode
• Improved performance of audio graph rendering

Version 0.12.0 (2021-12-08)

• Support all AudioParam automation methods
• Change AudioProcessor API to match spec better

Version 0.11.0 (2021-11-25)

• Support AudioParam.exponentialRampToValueAtTime
• Fix runaway memory and CPU usage due to unreleased nodes from graph

Version 0.10.0 (2021-11-10)

• Added IirFilterNode
• Added AudioContextOptions
• Support more audio mixing modes

Version 0.9.0 (2021-11-02)

• Added WaveShaperNode
• Added StereoPannerNode

Version 0.8.0 (2021-10-21)

• OscillatorNode improvements
• Added BiquadNode