Jan 29 · 6 min read · What is VAD (Voice Activity Detection)? Voice Activity Detection (VAD)isa digital signal processing technique that distinguishes human speech from background noise or silence. It acts as a "gatekeeper" in audio pipelines, ensuring systems only proces...
Join discussionJan 29 · 6 min read · What is VAD (Voice Activity Detection)? Voice Activity Detection (VAD) is a digital signal processing technique that distinguishes human speech from background noise or silence. It acts as a "gatekeeper" in audio pipelines, ensuring systems only proc...
Join discussionNov 11, 2025 · 3 min read · Meta has made a major return to open-source AI with its new Omnilingual ASR system—capable of transcribing speech in more than 1,600 languages, including over 500 that previously had no AI support. The platform also supports “Bring Your Own Language”...
Join discussion
Sep 13, 2025 · 5 min read · TL;DR The LLM4S Scala library has been extended to fully support speech-to-text (ASR) and text-to-speech (TTS) as first-class modalities. This GSoC 2025 contribution adds a comprehensive speech subsystem, enabling voice input/output alongside text. K...
Join discussion
Jul 25, 2025 · 6 min read · When Windows systems reboot unexpectedly, it can be challenging to determine the root cause. This newsletter provides comprehensive guidance on investigating these mysterious events using event logs, system files, and virtualization-specific tools. T...
Join discussion
Apr 4, 2025 · 26 min read · Rich Transcription Time Marked (RTTM) is a widely used, text-based format for annotating audio and video, representing results of speech recognition, speaker diarization, and related metadata. Developed by NIST in the early 2000s, RTTM files consist ...
Join discussion
Jan 2, 2024 · 13 min read · Introduction Creating an Automatic Speech Recognition (ASR) system using Wav2Vec 2.0 on E2E’s Cloud GPU server is a compelling endeavor that brings together cutting-edge technology and robust infrastructure. Leveraging the power of Wav2Vec 2.0, a sta...
Join discussion
Oct 18, 2023 · 7 min read · To give my Ai learning a context to ground into, I'm writing a funny little app I've called VoxPlan (in Python) which allows you to organise goals and tasks in a hierarchical tree and display them in an interactive GUI. I'm very interested in explori...
Join discussion
Apr 15, 2023 · 9 min read · Background With the development of unsupervised pre-training, exemplified by Wav2Vec 2.0 released in 2020, these models could learn directly from the raw audio without the need for human labels. So the raw training data could be scaled to 1 million h...
Join discussion
Mar 31, 2023 · 4 min read · jambonz supports many speech providers out of the box, but what if you want to use a speech provider for that is not currently supported? There is where the jambonz custom speech API comes in. The custom speech api requires jambonz 0.8.2 or above I...
Join discussion