Search Results for "vad"

Bailing is an open-source voice-dialogue assistant designed to deliver natural voice-based conversations by combining automatic speech recognition (ASR), voice activity detection (VAD), a large language model (LLM), and text-to-speech (TTS) in a single pipeline. Its goal is to offer a “voice-first” chat experience similar to what one might expect from a system like GPT-4o, but fully open and deployable by users. The project is modular: each core function — ASR, VAD, LLM, TTS — exists as a separately replaceable component, which allows flexibility in picking your preferred models depending on resources or languages. ...

RealtimeSTT is a Python-based realtime speech-to-text engine emphasizing low latency, wake-word detection, voice activity detection, and automatic speech segmentation. It provides asynchronous callbacks, nanosecond-precision timestamps, and CLI tools, suitable for building voice assistants, meeting transcribers, or live caption systems.

WhisperJAV is an open-source speech transcription pipeline designed specifically for generating subtitles for Japanese adult video content. The project addresses challenges that standard speech recognition models face when transcribing this type of audio, which often includes low signal-to-noise ratios and large numbers of non-verbal vocalizations. Traditional automatic speech recognition systems can misinterpret these sounds as words, leading to inaccurate transcripts. WhisperJAV introduces...

...Its backend leverages OpenAI’s Whisper models for GPU-accelerated speech recognition and Parakeet V3 for efficient CPU-only transcription with automatic language detection. To further refine accuracy and responsiveness, Handy integrates Silero’s Voice Activity Detection (VAD) for silence filtering, ensuring only speech segments are processed.

A subtitle generator for Japanese Adult Videos. Transformer-based ASR architectures like Whisper suffer significant performance degradation when applied to the spontaneous and noisy domain of JAV. This degradation is driven by specific acoustic and temporal characteristics that defy the statistical distributions of standard training data.

...Under the hood, Amica leverages modern web and desktop technologies: three.js and three-vrm for 3D rendering, Transformers.js for running models in the browser, Whisper and Silero VAD for speech recognition and voice-activity detection, and a variety of LLM backends such as llama.cpp servers, ChatGPT-compatible APIs, Ollama, KoboldCpp, and others. It also integrates multiple text-to-speech providers, including ElevenLabs, OpenAI, Coqui, RVC, and AllTalkTTS.