Repository navigation
Voice models and ten new media tools.
wiro-0.2.0.difypkg: the Wiro AI model provider. Adds 9 text to speech models (ElevenLabs, Google Gemini, Kokoro, Qwen3 TTS and MiMo) and 11 speech to text models (ElevenLabs, Whisper, NVIDIA Parakeet and Nemotron, Qwen3 ASR, Cohere Transcribe and others) for Dify's voice features, next to the 65 chat models of the gateway. Voices come from the Wiro AI catalog, and voice models can be added by ID.wiro_media-0.2.0.difypkg: the Wiro AI Media tools. New: Edit Image, Upscale Image, Remove Background, Image to Video, Lipsync, Speech to Text, Generate 3D, Run Workflow, Ask Studio and Message Agent, for 18 tools in all. Each model tool lists the catalog's models for its task, and every result includes its cost. Get Task Result also collects workflow runs.
Upgrading from 0.1.0 keeps your saved API keys. Install in Dify from Plugins > Install Plugin > Local Package File, or from the Dify Marketplace once 0.2.0 is listed there. Both need Dify 1.14.2 or later and a Wiro AI API key.