dsh-fish-tts

When using the DeepSeek Harness Web GUI for AI conversations, you might want to have the assistant's replies read aloud to enhance accessibility or multitasking. However, generic text-to-speech solutions often lack flexibility, such as using custom voices or managing additional APIs. dsh-fish-tts is a third-party TTS plugin designed to address this, leveraging the Fish Audio API so you can use your own API key to vocalize each reply with a personalized voice.

The plugin's core strength lies in its seamless integration and customization. It adds a 'Read aloud' button to every assistant message for instant playback, along with an 'auto-read' toggle in the input bar to automatically vocalize new responses. Through a settings page, you can configure TTS models, voice (including cloned voices), API key (securely encrypted with AES-256-GCM), and HTTP proxy. Additionally, a bilingual interface (Chinese/English) adapts to DSH's language settings, offering features like trial playback, volume, and speed adjustment.

dsh-fish-tts is ideal for users who already have a Fish Audio account and seek high-quality or personalized speech. If you're looking to move beyond default fixed voices and use your own trained or cloned voice to read AI replies, this plugin is a perfect fit. It emphasizes security (like local isolation and encrypted storage) and ease of use, allowing you to quickly set up and enjoy a tailored voice experience without deep technical involvement.