Fish Audio Integrations

VoiSpark is a browser-based AI voice generation platform that transforms text into natural, human-like speech across 30+ languages and dialects, offering over 100 voice templates spanning ages, accents, and personas. It supports real-time streaming with open source models like Nari Labs Dia and premium engines such as ElevenLabs, all accessible via a simple web interface or REST API. Users can fine-tune voice characteristics through intuitive sliders and context-aware generation that adapts pacing and tone to any script. Instant 30-second previews let you sample voices risk-free, while multi-format flexibility enables text input via typing, PDF uploads, or Google Docs syncing and exports as MP3 or WAV for seamless editing. Advanced features include voice cloning from short samples, switchable "professional” and “expressive” models for clarity or creativity, and batch generation for podcasts, e-learning, audiobooks, video dubbing, social media clips, and game character voices.

Vision Agents is an open source Python framework for building low-latency voice and video AI agents with any model. It lets developers plug in LLM, speech, and vision models from more than 25 providers and ship real-time agents for telehealth, voice support, live coaching, video analysis, interactive avatars, security monitoring, sports commentary, and other multimodal applications. It is designed to help teams build agents that can listen, speak, see, process media, call tools, and respond in real time while running on Stream’s global edge network with sub-500ms latency. Developers can build a first agent in minutes, using a small Python setup with Gemini Realtime, OpenAI, Deepgram, ElevenLabs, Stream, or other supported providers. Vision Agents supports both real-time speech-to-speech models and custom STT/LLM/TTS pipelines, giving teams either the fastest path to a working voice agent or full control over speech recognition, language reasoning, text-to-speech, etc.

VideoLingo is a hosted video localization service for course instructors, educators, technical creators, bloggers, and other publishers preparing videos for multilingual audiences. It generates bilingual subtitles and dubbed video, with translation designed to preserve cultural expressions and specialized terminology. Its workflow includes subtitle segmentation and timing, synthesized speech, and video export; features and limits vary by plan. The free plan provides 15 minutes of video processing, English and Chinese input, bilingual subtitles, 480p output, dubbing, and a watermark. Paid plans add processing capacity, support for eight input languages, 1080p output, and additional dubbing options. An open-source edition can also be installed locally on Windows, macOS, or Linux.

Zoice is a browser-based AI avatar video generator for creators, marketers, educators, and teams producing social, promotional, training, and explainer content. Users can upload a photo or create a character, add a script or audio, choose or import a voice, then render a talking or singing avatar video. The workflow includes lip synchronization, facial movement and gestures, action prompts, reusable avatar and voice profiles, speech preview, and aspect-ratio and resolution controls with outputs up to 4K. Zoice supports scripts in more than 170 languages, voice cloning, and importing voice profiles from services such as ElevenLabs and Fish Audio. A developer API also supports avatar-video generation. The service is delivered online and uses credit-based generation; avatar video generation is included in paid subscriptions.