Articles tagged "音声AI"

Irodori-TTS-v4-Large is a 3.29B Japanese-only TTS with voice cloning, Voice Design and emoji control. Setup, benchmarks vs v4.1-Small, Gemma license notes.

A guide to debpalash/VoiceStudio (formerly OmniVoice-Studio), which surged on GitHub Trending in September 2026. Covers how it runs voice cloning, voice design, video dubbing, dictation, transcription, and audiobook creation entirely on local hardware, its 16 TTS / 11 ASR engine lineup, installation, local API/MCP support, and how it compares to ElevenLabs and other tools.

GPT-Live-1 is OpenAI's full-duplex voice model, released Sep 10, 2026, billed $0.05/min on a dedicated Live endpoint, with 87% tool accuracy and 0.8s latency.

On May 7, 2026, OpenAI released a trio of new voice models: GPT-Realtime-2 (the first voice model with GPT-5-class reasoning), GPT-Realtime-Translate (live translation across 70+ input / 13 output languages), and GPT-Realtime-Whisper (streaming speech-to-text). This article summarizes capabilities, benchmark deltas vs 1.5, pricing, when to pick which, and the upgrade decision from 1.5 — based on official information.

OpenAI released the gpt-realtime-1.5 audio model on February 26, 2026, and openai/realtime-voice-component on GitHub provides an official React reference for voice UIs. This article summarizes the documented gains (+5% audio reasoning, +10.23% transcription, +7% instruction following), pricing, the component's positioning as a reference implementation, and practical considerations for business adoption.

NVIDIA PersonaPlex 7B, released in January 2026, is an open-source voice AI that integrates the traditional ASR→LLM→TTS pipeline into a single end-to-end model, achieving true full-duplex voice interaction. This guide covers architecture, performance benchmarks, setup procedures, and practical use cases.