High Performance Web Audio with AudioWorklet in Firefox
Earlier this week, Audio Worklets landed in the release of Firefox 76. We’re ready to start bridging the gap between web audio and native. Developers can now leverage
AudioWorkletto write arbitrary audio processing code. This exciting new functionality raises the bar for emerging web experiences like 3D games, VR, and music production.
DeepSpeech 0.6: Mozilla’s Speech-to-Text Engine Gets Fast, Lean, and Ubiquitous
The Machine Learning team at Mozilla continues work on DeepSpeech, an automatic speech recognition (ASR) engine which aims to make speech recognition technology and trained models openly available to developers. In this overview of recent improvements, we'll show how DeepSpeech can transform your applications by enabling client-side, low-latency, and privacy-preserving speech recognition capabilities. Find out how you can participate.
A Real-Time Wideband Neural Vocoder at 1.6 kb/s Using LPCNet
This is an update on the LPCNet project, an efficient neural speech synthesizer from Mozilla’s Emerging Technologies group. LPCNet combines signal processing and deep learning to improve the efficiency of neural speech synthesis. Our recent work turns LPCNet into a very low-bitrate neural speech codec that’s actually usable on current hardware and even on phones.
Firefox 66 to block automatically playing audible video and audio
Unsolicited volume can be a great source of distraction and frustration for users of the web. So we are making changes to how Firefox handles playing media with sound and we want to make sure web developers are aware of this new audio autoplay blocking default. With the release of Firefox 66, now in Firefox Beta/Developer Edition, the browser will block audible audio and video, and will allow a site to play audio or video aloud via the
HTMLMediaElementAPI only once the user has initiated the audio.
LPCNet: DSP-Boosted Neural Speech Synthesis
LPCNet is a new project out of Mozilla’s Emerging Technologies group — an efficient neural speech synthesiser with reduced complexity over some of its predecessors. Neural speech synthesis models have already demonstrated impressive speech synthesis quality, but their computational complexity has made them hard to use in real-time, especially on phones. Our solution with LPCNet uses a combination of deep learning and digital signal processing (DSP) techniques.
Introducing Opus 1.3
Opus is a totally open, royalty-free, audio codec that can be used for all audio applications, from music streaming and storage to high-quality video-conferencing and VoIP. This 1.3 release brings quality improvements to both speech and music compression, ambisonics support, and more.
RNNoise: Using Deep Learning for Noise Suppression
The Mozilla Research RRNoise project shows how to apply deep learning to noise suppression. It combines classic signal processing with deep learning, but it’s small and fast. And you can help! Find out how to donate your noise to science.
Opus audio codec version 1.2 released
The Opus audio codec just got another major upgrade with the release of version 1.2, bringing many speech and music quality improvements, especially at low bitrates.
What’s new in Web Audio?
The Web Audio API is still under development, which means there are new methods and properties being added, renamed, shuffled around or simply removed! In this article, we look at what’s happened since our last update in early 2015, both in the Web Audio specification and in Firefox’s implementation. The demos all work in Firefox […]
Record almost everything in the browser with MediaRecorder
The MediaRecorder API lets you record media streams, i.e. moving images and audio. The result of these recordings can be, for example, an OGG file, like the ones you use to listen to music. Browser-wise, we can obtain streams in many ways. Let’s start with something you might be familiar with: we’ll get a stream […]