Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Binary file added public/logos/scowld.png
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
129 changes: 129 additions & 0 deletions src/content/tools/scowld.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,129 @@
---
name: Scowld
slug: scowld
tagline: Open-source iOS voice companion with BYOK AI, STT, TTS, and vision
website: 'https://scowld.xyz'
logo: /logos/scowld.png
primary_category: voice_agent_platform
secondary_categories:
- tts
- stt
layers:
- L3
use_cases:
creators:
- vtuber
game_devs: []
voice_ai_builders:
- virtual_companion
- voice_assistant
- conversational_agent
pricing:
model: open_source
has_free_tier: true
starting_paid_usd: 0
pricing_url: 'https://github.com/apoorvdarshan/scowld'
licensing:
commercial_use: true
youtube_monetization: true
game_use: true
voice_cloning_allowed: false
notes: >-
Scowld's code is MIT licensed. AI, speech-to-text, text-to-speech, model,
voice, and generated-output rights remain subject to the terms of each
provider configured by the user.
capabilities:
voice_cloning: false
multilingual: true
chinese_support: false
realtime_capable: false
open_source: true
offline_capable: false
batch_api: false
gotchas:
- >-
The app includes no bundled provider credits; users must configure their
own AI and speech provider keys
- >-
Chat, cloud transcription, optional vision, and generated speech are sent
directly to the providers selected by the user
- >-
Hands-free wake detection is local, but complete AI conversations require
network access to the configured cloud providers
portability:
voice_model_export: false
notes: >-
Chats and provider settings are local to the device, while model and voice
portability depend on the external providers the user selects.
alternatives:
- openvoiceos
- pipecat
verified_at: 2026-08-07
badges:
- open_source
- new
i18n:
zh:
tagline: 支持自备密钥 AI、语音识别、语音合成和视觉的开源 iOS 语音伙伴
licensing_notes: Scowld 代码采用 MIT 许可;AI、语音识别、语音合成、模型、声音及生成内容的权利仍取决于用户所选服务商的条款。
gotchas:
- 应用不提供内置服务额度,用户必须配置自己的 AI 和语音服务密钥
- 聊天、云端转写、可选视觉内容和生成语音会直接发送到用户选择的服务商
- 唤醒词检测在设备端完成,但完整 AI 对话仍需要连接所配置的云端服务
portability_notes: 聊天记录和服务商设置保存在设备本地;模型和声音能否迁移取决于用户选择的外部服务商。
body: >-
## 先判断它是否适合你的使用方式

Scowld 是一款面向 iPhone 和 iPad 的开源语音伙伴。它把动画角色、语音或文字聊天、可选视觉、历史对话以及自备密钥服务商整合在一个应用中,适合希望直接体验虚拟伙伴或研究语音交互流程的人。

## 使用前先准备服务商密钥

应用本身免费且没有订阅,但不会赠送 AI 或语音额度。用户需要为聊天、云端语音识别和语音合成配置自己的服务商账号与密钥,并自行承担对应费用和限制。

## 注意隐私与授权边界

密钥保存在 iOS 钥匙串中,聊天记录保存在设备本地;发送消息、音频、可选图像和生成语音时,数据会直接交给用户选择的服务商。MIT 许可覆盖 Scowld 代码,但具体模型、声音和输出内容仍要遵守各服务商条款。
github_metrics:
repo: apoorvdarshan/scowld
source_url: 'https://github.com/apoorvdarshan/scowld'
captured_at: 2026-08-07
stars: 11
forks: 2
watchers: 0
open_issues: 0
open_pull_requests: 0
last_commit_at: 2026-07-25T10:26:53Z
latest_release_at: 2026-07-09T10:31:47Z
has_releases: true
license: MIT
primary_language: Swift
archived: false
signals:
- maintained
note: >-
Public GitHub metrics captured for maintenance screening; verify repository
activity before adopting it for production.
---
## Decide whether Scowld fits your workflow

Scowld is an open-source iOS voice companion rather than a hosted voice-agent platform. It combines an animated VRM character with voice and text chat, optional camera context, saved conversations, and user-selected AI, STT, and TTS providers.

### Use it for a customizable companion on iPhone or iPad

It is a practical fit when you want a ready-made mobile companion experience while retaining control over provider selection and API keys. VTuber creators and voice-AI builders can also use it as a concrete reference for combining a character, wake mode, transcription, model responses, and spoken output.

### Avoid it when you need a managed service

Scowld does not include hosted credits, a managed backend, telephony, batch generation, or a browser-based production console. It is not the shortest route for teams that want a turnkey commercial voice-agent service.

## Plan the provider setup first

The app is free and has no subscription, but the configured AI, STT, and TTS providers may charge separately. A complete spoken conversation normally requires an AI provider key plus an ElevenLabs or OpenAI text-to-speech key.

### Keep provider terms with the project

The MIT license permits commercial use of Scowld's code. Model access, generated output, reference voices, and any provider-specific commercial rights remain governed by the chosen provider, so recheck those terms before publishing or shipping paid work.

## Understand the privacy boundary

Provider keys are stored in iOS Keychain and chat history stays on the device. Hands-free wake detection runs locally while enabled, but messages, recorded speech, optional camera context, and generated speech are processed by the cloud services the user selects.