Search Results for "private-gpt"

...For voice, it integrates with numerous TTS engines (Edge-TTS, VITS-Fast, ElevenLabs, VALL-E-X, OpenVoice, GPT-SoVITS, Azure TTS, fish-speech, ChatTTS, CosyVoice, F5-TTS, MultiTTS, MeloTTS, and others), and can optionally pass the output through voice conversion systems like so-vits-svc or DDSP-SVC to change timbre.

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. It aims to be light enough to run without a GPU, making it usable on modest hardware or edge devices, while still maintaining low latency and smooth interaction. ...

...The system is customizable: you can define your own avatar appearance and voice, and it supports voice cloning so you can generate a new voice from a short 3–10 second reference sample. The tech stack integrates FunASR for speech recognition, Qwen for language understanding, multiple TTS engines like GPT-SoVITS, CosyVoice, or edge-tts, and MuseTalk for talking-head generation.

...It is model-agnostic and advertises support for a variety of TTS and speech models such as ChatTTS, CosyVoice, Fish-Speech, FireredTTS and others, as well as Whisper-based ASR, giving you a flexible playground for experimenting with different speech stacks. The project also integrates with general-purpose LLMs (for example GPT- or LLaMA-style models), which can be used to pre-process text, manage conversations.

WavTokenizer is a state-of-the-art discrete acoustic codec designed specifically for audio language modeling, capable of compressing 24 kHz audio into just 40 or 75 tokens per second while preserving high perceptual quality. It is built to represent speech, music, and general audio with extremely low bitrate, making it ideal as a front-end for large audio language models like GPT-4o and similar architectures. The model uses a single-quantizer design together with temporal compression to achieve extreme compression without sacrificing reconstruction fidelity. Its architecture incorporates a broader vector-quantization space, extended contextual windows, and improved attention networks, combined with multi-scale discriminators and inverse Fourier transform blocks to enhance waveform reconstruction. ...