jing332/tts-server-android
Tts Server Android
tts-server-android este un serviciu de text-to-speech la nivel de sistem pentru Android care direcționează cererile de sinteză către API-uri cloud externe sau motoare locale. Funcționează ca un gateway HTTP de sinteză vocală, convertind cererile vocale ale sistemului în cereri HTTP customizabile pentru servicii cloud remote.
Proiectul include un parser de dialog narativ care utilizează ghilimele pentru a diferenția între narațiune și dialog, permițând stiluri diferite de citire. Dispune, de asemenea, de un manager de voce și o interfață de sinteză pentru a implementa reguli de înlocuire a textului și reîncercări automate pentru a îmbunătăți acuratețea output-ului vocal.
Serviciul acoperă capabilități mai largi, inclusiv gestionarea motorului local pentru funcționalitatea vocală offline, rutarea API-urilor cloud și personalizarea pronunției prin preprocesarea textului bazată pe reguli.
Features
- Speech Synthesis Gateways - Functions as a gateway that routes system speech requests to external cloud TTS providers via HTTP.
- System TTS Integrations - Implements a system-level service provider that integrates with the Android OS to route text synthesis requests.
- Text-to-Speech Integrations - Routes synthesis requests to external cloud services or local engines to deliver device-wide speech functionality.
- Pronunciation Rules - Implements a framework for text replacement rules and automatic retries to ensure accurate voice output.
- Text-to-Speech Synthesis - Processes input text through replacement rules and retries to ensure accurate speech generation.
- Voice Management - Includes a voice manager to customize audio output through styles and pronunciation configurations.
- Local Speech Synthesis - Loads external speech synthesis engines to provide offline text-to-speech functionality on the device.
- Local Engine Plugins - Loads and integrates third-party speech synthesis libraries to enable offline text-to-speech capabilities.
- Text Preprocessing - Modifies raw input text using replacement rules to ensure correct pronunciation before synthesis.
- Local Engine Management - Imports and manages third-party speech synthesis engines to provide offline voice capabilities on mobile devices.
- Cloud Speech Synthesizers - Routes synthesis requests to external cloud services using custom HTTP configurations for high-quality output.
- HTTP Request Routing - Routes text requests to remote cloud servers using customizable HTTP configurations to retrieve audio streams.
- Text-to-Speech Normalizers - Adjusts pronunciation rules and applies text replacement to improve the naturalness of synthesized speech.
- Speech Synthesis Providers - Connects external TTS APIs and local engines to serve as the default provider for system-wide voice output.
- Dialogue Analysis - Provides a parser that uses quotation marks to differentiate between narration and dialogue for distinct reading styles.
- Dialogue and Narrative Parsers - Uses quotation marks to automatically separate narrative voice from character dialogue for different reading styles.
- TTS-Specific Configurations - Defines pronunciation rules and voice settings to customize how text is read aloud.
- Provider Failover Mechanisms - Provides logic to automatically switch to backup speech providers when the primary API fails.
- Voice Character Presets - Dynamically changes voice styles and characters using specific pronunciation configurations.
