Refactor

[basnijholt]

No description provided.

[cursor]

Bug: ASR Interface Refactor Causes Parameter Mismatch

The ASR interface was refactored from a function to an ASRService object. In agent_cli/agents/chat.py and agent_cli/agents/transcribe.py, the transcriber.transcribe() method is called with parameters like p, stop_event, quiet, live, and logger. However, the ASRService.transcribe() method, as defined in its base class, only accepts audio_data: bytes, leading to a TypeError at runtime.

agent_cli/agents/chat.py#L178-L193

agent-cli/agent_cli/agents/chat.py

Lines 178 to 193 in 3d0c3b9

	start_time = time.monotonic()
	transcriber = get_asr_service(
	provider_cfg,
	wyoming_asr_cfg,
	openai_asr_cfg,
	openai_llm_cfg,
	LOGGER,
	quiet=general_cfg.quiet,
	)
	instruction = await transcriber.transcribe(
	p=p,
	stop_event=stop_event,
	quiet=general_cfg.quiet,
	live=live,
	logger=LOGGER,
	)

agent_cli/agents/transcribe.py#L84-L99

agent-cli/agent_cli/agents/transcribe.py

Lines 84 to 99 in 3d0c3b9

	with signal_handling_context(LOGGER, general_cfg.quiet) as stop_event:
	transcriber = get_asr_service(
	provider_cfg,
	wyoming_asr_cfg,
	openai_asr_cfg,
	openai_llm_cfg,
	LOGGER,
	quiet=general_cfg.quiet,
	)
	transcript = await transcriber.transcribe(
	logger=LOGGER,
	p=p,
	stop_event=stop_event,
	quiet=general_cfg.quiet,
	live=live,
	)

Fix in Cursor • Fix in Web

---

Bug: Function Call Error: Mismatched Parameters

Interface mismatch: The synthesizer function returned by get_synthesizer() now only accepts a text parameter. However, it is still being called with additional keyword arguments such as wyoming_tts_config, openai_tts_config, openai_llm_config, logger, quiet, and live, which will cause a TypeError at runtime.

agent_cli/tts.py#L287-L297

agent-cli/agent_cli/tts.py

Lines 287 to 297 in 3d0c3b9

	async with live_timer(live, "🔊 Synthesizing text", style="blue", quiet=quiet):
	audio_data = await synthesizer(
	text=text,
	wyoming_tts_config=wyoming_tts_config,
	openai_tts_config=openai_tts_config,
	openai_llm_config=openai_llm_config,
	logger=logger,
	quiet=quiet,
	live=live,
	)
	except Exception:

Fix in Cursor • Fix in Web

---

Bug: ASR Service Fails Without Audio Input Configuration

The audio_in_cfg parameter, which configured the audio input device (e.g., input_device_index via setup_devices) for the ASR service, was removed from the _handle_conversation_turn and _async_main functions. This prevents the ASR service from correctly selecting the audio input device, causing transcription failures.

agent_cli/agents/transcribe.py#L69-L79

agent-cli/agent_cli/agents/transcribe.py

Lines 69 to 79 in 3d0c3b9

	async def _async_main(
	*,
	provider_cfg: config.ProviderSelection,
	general_cfg: config.General,
	wyoming_asr_cfg: config.WyomingASR,
	openai_asr_cfg: config.OpenAIASR,
	ollama_cfg: config.Ollama,
	openai_llm_cfg: config.OpenAILLM,
	llm_enabled: bool,
	p: pyaudio.PyAudio,

agent_cli/agents/chat.py#L145-L162

agent-cli/agent_cli/agents/chat.py

Lines 145 to 162 in 3d0c3b9

	async def _handle_conversation_turn(
	*,
	p: pyaudio.PyAudio,
	stop_event: InteractiveStopEvent,
	conversation_history: list[ConversationEntry],
	provider_cfg: config.ProviderSelection,
	general_cfg: config.General,
	history_cfg: config.History,
	wyoming_asr_cfg: config.WyomingASR,
	openai_asr_cfg: config.OpenAIASR,
	ollama_cfg: config.Ollama,
	openai_llm_cfg: config.OpenAILLM,
	audio_out_cfg: config.AudioOutput,
	wyoming_tts_cfg: config.WyomingTTS,
	openai_tts_cfg: config.OpenAITTS,
	live: Live,
	) -> None:

Fix in Cursor • Fix in Web

---

BugBot free trial expires on July 22, 2025
You have used $0.00 of your $50.00 spend limit so far. Manage your spend limit in the Cursor dashboard.

Was this report helpful? Give feedback by reacting with 👍 or 👎