Add TTS support, Discord thread fix, and test updates by loadxf · Pull Request #34 · loadxf/Gwenn_ai

loadxf · 2026-03-02T00:29:57Z

Add TextToSpeechSynthesizer (ElevenLabs) with Discord/Telegram integration
Fix Discord thread parent_id resolution for session scoping
Update video media module and config for new media features
Update tests for daemon, MCP client, orchestration, and coverage gaps

- Add TextToSpeechSynthesizer (ElevenLabs) with Discord/Telegram integration - Fix Discord thread parent_id resolution for session scoping - Update video media module and config for new media features - Update tests for daemon, MCP client, orchestration, and coverage gaps Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Copilot

Pull request overview

Adds optional ElevenLabs-based text-to-speech (TTS) support across CLI, Telegram, and Discord, while also fixing Discord thread session scoping and tightening tests/coverage to reduce unawaited-coroutine warnings.

Changes:

Introduce TextToSpeechSynthesizer (ElevenLabs) and integrate voice replies/playback in Discord, Telegram, and CLI.
Fix Discord thread parent_id handling so session scope keys resolve to the parent channel while still tracking the thread id.
Update tests and pytest configuration to address resource warnings / unawaited coroutine warnings and close leaked resources.

Reviewed changes

Copilot reviewed 15 out of 15 changed files in this pull request and generated 5 comments.

Show a summary per file

File	Description
`gwenn/media/tts.py`	New ElevenLabs TTS synthesizer with lazy client init + graceful fallback behavior.
`gwenn/media/__init__.py`	Exports `TextToSpeechSynthesizer` from `gwenn.media`.
`gwenn/channels/telegram_channel.py`	Adds optional TTS voice replies (mode-dependent) and closes synthesizer on shutdown.
`gwenn/channels/discord_channel.py`	Adds optional TTS voice replies and fixes thread parent channel scoping for sessions.
`gwenn/main.py`	Adds CLI TTS playback and updates startup/daemon UI messaging text.
`gwenn/config.py`	Adds `ElevenLabsConfig` and wires it into the main config.
`gwenn/media/video.py`	Ensures `cv2.VideoCapture.release()` always happens via `try/finally`.
`pyproject.toml`	Adds `elevenlabs` extra; configures pytest asyncio strict mode and warning filters.
`tests/test_tts.py`	New unit tests for the TTS synthesizer (mocked ElevenLabs client).
`tests/test_daemon.py`	Ensures client disconnect happens to prevent StreamWriter resource warnings.
`tests/test_mcp_client.py`	Closes unawaited coroutine objects passed to mocked `create_task`.
`tests/test_orchestration_coverage.py`	Closes unawaited coroutine objects passed to mocked `ensure_future`.
`tests/test_remaining_coverage.py`	Adds an autouse fixture to close `MemoryStore` instances created during tests.
`tests/test_main_coverage.py`	Adjusts asyncio.run mocking to close coroutines (but reduces assertions).
`tests/test_final_coverage_gaps.py`	Avoids creating unawaited coroutine in semaphore acquisition simulation (via internal `_value`).

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-03-02T00:35:26Z

+        # Acquire the semaphore first so release() works.
+        # _sync_slot is an asyncio.BoundedSemaphore — decrement its value
+        # directly to simulate acquisition without creating an unawaited coroutine.
+        executor._sync_slot._value -= 1


This test now uses the private asyncio.BoundedSemaphore implementation detail _value to simulate an acquisition. That’s brittle across Python versions/asyncio implementations and can make the test fail unexpectedly if the semaphore internals change. Prefer acquiring via a real event loop (e.g., asyncio.run(executor._sync_slot.acquire())) or otherwise structuring the test so it doesn’t depend on _value internals.

Suggested change

# Acquire the semaphore first so release() works.

# _sync_slot is an asyncio.BoundedSemaphore — decrement its value

# directly to simulate acquisition without creating an unawaited coroutine.

executor._sync_slot._value -= 1

# Acquire the semaphore first so release() works, using the public API.

# Use a real event loop instead of mutating the private _value attribute.

asyncio.run(executor._sync_slot.acquire())

Copilot · 2026-03-02T00:35:26Z

    def test_main_default_runs_session(self, monkeypatch):
        from gwenn.main import main
        monkeypatch.setattr("sys.argv", ["gwenn"])
-        mock_run = MagicMock()
-        monkeypatch.setattr("asyncio.run", mock_run)
+
+        def _mock_run(coro):
+            coro.close()
+
+        monkeypatch.setattr("asyncio.run", _mock_run)
        monkeypatch.setattr("gwenn.main._logging_configured", False)
        monkeypatch.setattr("gwenn.main._termios", None)
        main()


These entry-point tests no longer assert that asyncio.run(...) was invoked; they only close the coroutine to avoid warnings. That weakens the test because it would still pass if main() stopped running the session entirely. Consider keeping an assertion by wrapping asyncio.run with a stub that records call count/args (and still closes the passed coroutine).

Copilot · 2026-03-02T00:35:27Z

            self._set_startup_step(daemon_state, "connect", "done")
            daemon_state["ready_lines"] = [
-                "Connected to Gwenn's daemon. Type your message, or '/exit' to close.",
+                "Connected to Gwenn's framework. Type your message, or '/exit' to close.",


The daemon-connection success message says "Connected to Gwenn's framework", but this code path is specifically connecting to the daemon socket (CliChannel.connect). This is likely to confuse users troubleshooting daemon connectivity; consider changing this line back to "Connected to Gwenn's daemon" (or similar) to match the actual behavior.

Suggested change

"Connected to Gwenn's framework. Type your message, or '/exit' to close.",

"Connected to Gwenn's daemon. Type your message, or '/exit' to close.",

Copilot · 2026-03-02T00:35:27Z

+class ElevenLabsConfig(BaseSettings):
+    """Configuration for ElevenLabs text-to-speech (optional)."""
+
+    api_key: Optional[str] = Field(None, alias="ELEVENLABS_API_KEY")
+    voice_id: str = Field("JBFqnCBsd6RMkjVDRZzb", alias="GWENN_TTS_VOICE_ID")
+    model: str = Field("eleven_turbo_v2_5", alias="GWENN_TTS_MODEL")
+    output_format: str = Field("opus_48000_128", alias="GWENN_TTS_OUTPUT_FORMAT")
+    max_chars: int = Field(2500, alias="GWENN_TTS_MAX_CHARS")
+    mode: str = Field("voice_reply", alias="GWENN_TTS_MODE")
+
+    model_config = {"env_file": _ENV_FILE, "extra": "ignore"}
+
+    @property
+    def is_available(self) -> bool:
+        return bool(self.api_key)
+
+    def should_send_voice(self, is_voice_message: bool = False) -> bool:
+        """Whether a voice reply should be sent for this interaction."""
+        if not self.is_available or self.mode == "off":
+            return False
+        if self.mode == "always":
+            return True
+        return self.mode == "voice_reply" and is_voice_message
+
+    @model_validator(mode="after")
+    def validate_mode(self) -> "ElevenLabsConfig":
+        if self.mode not in {"off", "voice_reply", "always"}:
+            import structlog
+            structlog.get_logger(__name__).warning(
+                "config.elevenlabs_invalid_mode",
+                provided=self.mode,
+                fallback="voice_reply",
+            )
+            self.mode = "voice_reply"
+        return self


New ElevenLabs / TTS environment variables are introduced here (e.g., ELEVENLABS_API_KEY, GWENN_TTS_*), but they aren’t documented in the existing configuration reference or sample env file. Please update docs/configuration.md and .env.example to include these new settings so users can discover and configure TTS correctly.

Copilot · 2026-03-02T00:35:27Z

+[tool.pytest.ini_options]
+asyncio_mode = "strict"
+filterwarnings = [
+    # PTB plans to change retry_after to timedelta in a future major version.
+    # Nothing to act on until that release; silence the noise.
+    "ignore::telegram.warnings.PTBDeprecationWarning",
+    # unittest.mock creates unawaited coroutines when AsyncMock attributes are
+    # accessed but never called — a known CPython mock limitation, not a bug in
+    # our code.  Also covers coroutines passed to mocked ensure_future/create_task.
+    "ignore:coroutine 'AsyncMockMixin._execute_mock_call' was never awaited:RuntimeWarning",
+    "ignore:coroutine 'DockerManager.run_container.<locals>._cleanup_after_exit' was never awaited:RuntimeWarning",
+    # Rich animation internals may create unawaited sleep coroutines when a
+    # console status/spinner is interrupted by an exception.
+    "ignore:coroutine 'sleep' was never awaited:RuntimeWarning",
+    # structlog processor-chain ordering advisory — not actionable without
+    # removing pretty-exception support.
+    "ignore:Remove `format_exc_info` from your processor chain:UserWarning",
+]


The warning filter ignore:coroutine 'sleep' was never awaited:RuntimeWarning is very broad and will suppress any unawaited sleep() coroutine warnings across the entire test suite (including ones introduced by this repo). Consider narrowing this filter to the specific module(s) that emit it (e.g., Rich internals) so genuine unawaited coroutine regressions aren’t hidden.

Copilot AI review requested due to automatic review settings March 2, 2026 00:29

loadxf merged commit 984b1aa into main Mar 2, 2026
4 of 8 checks passed

Copilot started reviewing on behalf of loadxf March 2, 2026 00:30 View session

Copilot AI reviewed Mar 2, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add TTS support, Discord thread fix, and test updates#34

Add TTS support, Discord thread fix, and test updates#34
loadxf merged 1 commit into
mainfrom
dev

loadxf commented Mar 2, 2026

Uh oh!

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Mar 2, 2026

Uh oh!

Copilot AI Mar 2, 2026

Uh oh!

Copilot AI Mar 2, 2026

Uh oh!

Copilot AI Mar 2, 2026

Uh oh!

Copilot AI Mar 2, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

	"Connected to Gwenn's framework. Type your message, or '/exit' to close.",
	"Connected to Gwenn's daemon. Type your message, or '/exit' to close.",

Conversation

loadxf commented Mar 2, 2026

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 2, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants