feat(tts): implement tts

[qbc2016]

AgentScope Version

1.0.9dev

Description

As the title says.

Checklist

Please check the following items before code is ready to be reviewed.

• Code has been formatted with pre-commit run --all-files command
• All tests are passing
• Docstrings are in Google style
• Related documentation has been updated (e.g. links, examples, etc.)
• Code is ready for review

[DavdGao]

plz see inline comments

[cla-assistant]

All committers have signed the CLA.

[DavdGao]

@qbc2016

1. Please test more model and voice combinations, e.g. the combination of RealtimeTTSModel * qwen3-tts-flash-realtime * Serena doesn't work at all.
2. When we ask the agent to generate a long response (e.g. 背诵长恨歌）, the realtime API will be blocked for unknown reason in the example.

[Copilot]

Pull request overview

This PR implements comprehensive Text-to-Speech (TTS) functionality for AgentScope, introducing a unified interface for multiple TTS API providers including OpenAI, Gemini, and DashScope. The implementation supports both realtime (streaming input) and non-realtime TTS models, with seamless integration into the ReActAgent class.

Key Changes:

• Added TTS base class and model implementations for 4 different TTS APIs with streaming support
• Integrated TTS capability into ReActAgent with automatic speech synthesis
• Extended the Message class with a speech field for audio content

Reviewed changes

Copilot reviewed 23 out of 23 changed files in this pull request and generated 24 comments.

Show a summary per file

File	Description
src/agentscope/tts/_tts_base.py	Base class defining the TTS model interface with support for both realtime and non-realtime models
src/agentscope/tts/_tts_response.py	Data structures for TTS responses and usage tracking
src/agentscope/tts/_openai_tts_model.py	OpenAI TTS implementation with streaming output support
src/agentscope/tts/_gemini_tts_model.py	Gemini TTS implementation with streaming output support
src/agentscope/tts/_dashscope_tts_model.py	DashScope non-realtime TTS implementation
src/agentscope/tts/_dashscope_realtime_tts_model.py	DashScope realtime TTS with streaming input and output support
src/agentscope/tts/__init__.py	TTS module exports and public API
src/agentscope/agent/_react_agent.py	Integration of TTS into ReActAgent for automatic speech synthesis
src/agentscope/agent/_agent_base.py	Updated agent base to handle audio playback from speech field
src/agentscope/agent/_utils.py	Added async null context manager utility for optional TTS
src/agentscope/message/_message_base.py	Added speech field to Message class and updated get_text_content method
src/agentscope/hooks/_studio_hooks.py	Updated hook field names for studio integration
tests/tts_openai_test.py	Comprehensive unit tests for OpenAI TTS model
tests/tts_gemini_test.py	Comprehensive unit tests for Gemini TTS model
tests/tts_dashscope_test.py	Comprehensive unit tests for both DashScope TTS models
examples/functionality/tts/main.py	Example demonstrating TTS integration with ReActAgent
examples/functionality/tts/README.md	Documentation for the TTS example
docs/tutorial/en/src/task_tts.py	English tutorial for using TTS in AgentScope
docs/tutorial/zh_CN/src/task_tts.py	Chinese tutorial for using TTS in AgentScope
docs/tutorial/en/index.rst	Updated English documentation index
docs/tutorial/zh_CN/index.rst	Updated Chinese documentation index
README.md	Added TTS feature announcement
README_zh.md	Added TTS feature announcement in Chinese

[Copilot]

Pull request overview

Copilot reviewed 24 out of 24 changed files in this pull request and generated 20 comments.