Eliminate custom _emit_stt_ttfb_metric and manual timestamp tracking in
STTService by reusing FrameProcessor's start_ttfb_metrics/stop_ttfb_metrics
with new start_time/end_time parameters. This keeps the chronological
start→stop ordering and removes _speech_end_time and _last_transcription_time
state from STTService.
Remove the deprecation warning and __post_init__ override. Also fix the
default value for remote_participants to use field(default_factory=dict)
instead of None.
Add write_transport_frame() hook to BaseOutputTransport so subclasses
can handle custom frame types that flow through the audio queue. Add
DailySIPTransferFrame and DailySIPReferFrame as DataFrame subclasses
that queue with audio, ensuring SIP operations execute only after the
bot finishes its current utterance. Override write_transport_frame in
DailyOutputTransport to dispatch these frames to the existing
sip_call_transfer() and sip_refer() client methods.
Also switch DailyOutputTransport.send_message error handling from
logger.error to push_error for consistency.
RTVIObserver previously filtered out all upstream frames to avoid
duplicate messages from broadcasted frames. This caused upstream-only
frames to be silently ignored. Instead, add a `broadcasted` field to
the Frame base class that is set by broadcast_frame() and
broadcast_frame_instance(), and only skip upstream copies of
broadcasted frames.
Use client_req_id-based multiplexing instead of disconnecting and
reconnecting the websocket on every interruption. This follows the
same pattern used by Cartesia, ElevenLabs, and other services via
AudioContextWordTTSService.
Key changes:
- Base class: InterruptibleWordTTSService -> AudioContextWordTTSService
- Add close_ws_on_eos: False to setup message to keep connection alive
- Add client_req_id to text, end_of_stream messages for demultiplexing
- Route audio via append_to_audio_context() instead of push_frame()
- Silently drop messages for cancelled/unknown contexts on interruption
- Add _handle_interruption() that resets context without reconnecting
- Remove no-op push_frame() override
Always create UserIdleController (timeout=0 means disabled), removing
all Optional guards. Add UserIdleTimeoutUpdateFrame to allow changing
the idle timeout at runtime.
Replace the continuous heartbeat-based timer (UserSpeakingFrame/BotSpeakingFrame
+ asyncio.Event loop) with a simple one-shot timer that starts when
BotStoppedSpeakingFrame is received and cancels on UserStartedSpeakingFrame or
BotStartedSpeakingFrame. This eliminates false idle triggers caused by gaps
between the user finishing speaking and the bot starting to speak (LLM/TTS
latency).
Guard the timer start with two conditions to prevent false triggers:
- User turn in progress: during interruptions, BotStoppedSpeaking arrives
while the user is still speaking mid-turn.
- Function calls in progress: FunctionCallsStarted arrives before
BotStoppedSpeaking because the bot speaks concurrently with the function
call starting, so the timer must wait for the result and subsequent bot
response.
The outer try/except in each service decorator caught both tracing
setup errors and application errors from the wrapped function. If the
function itself raised (e.g. LLM rate limit, TTS timeout), the
exception was caught and the function was called a second time.
Fix by tracking whether the original function was called via a
fn_called flag. If the function was already called, re-raise the
exception instead of falling back to untraced re-execution.
The class_decorators.py module (Traceable, @traceable, @traced) is not
used anywhere in the codebase. Mark it deprecated and fix the misleading
comment in service_decorators.py that referenced it as if it were active.
Consolidate _tracing_enabled and _tracing_context from LLMService,
STTService, and TTSService into the shared AIService base class.
Extract _get_turn_context() helper in service_decorators.py to
encapsulate the repeated pattern across all traced decorators.
Add model-specific params (arcana: repetition_penalty, temperature, top_p;
mistv2: no_text_normalization, save_oovs, segment) with dynamic query param
building via _build_settings(). Model/voice/param changes now trigger
WebSocket reconnection since all settings are URL query params.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This allows non-websocket STT services (like SarvamSTTService, which uses
the Sarvam Python SDK for connection management) to reuse the same keepalive
pattern. Subclasses override _send_keepalive() and _is_keepalive_ready() for
their specific protocol.
