Deepgram query params

42 params recognised on the Deepgram session (/v1/listen with a DG credential). Rendered live from app/api/v1/_dg_params.py.

SUPPORTED Supported (10)

Param	Behaviour	Note
`encoding`	SUPPORTED	Parakeet expects PCM-16 (linear16); transcode mulaw/alaw/opus.
`sample_rate`	SUPPORTED	Parakeet trained at 16 kHz; resample any declared rate.
`endpointing`	SUPPORTED	Silero/WebRTC VAD; emit speech_final after silence>endpointing ms.
`language`	SUPPORTED	Parakeet-TDT-0.6B-v3 supports 25 languages; pass as model hint.
`numerals`	SUPPORTED	Parakeet TDT head includes ITN; enable via model config.
`keywords`	SUPPORTED	POST-TRANSCRIPTION ONLY: repeatable 'kw[:intensifier]' is parsed, stored on the session, and used to drive phonetic + fuzzy correction of the FINAL transcript against the locale medical lexicon (Cologne Phonetics for de, Double Metaphone for en/fr/es). Interim results are NEVER rewritten, so live captions do not flicker. This is NOT acoustic/decoder-level biasing: the decode is byte-identical with or without it, because the service stays pinned to decoding strategy 'greedy' to dodge an NVRTC CUDA-graph capture bug (services/nemo_decoding.py). A 2026-08-03 spike found that pin's blocker gone but did NOT demonstrate a boosting tree loading (fusion_models stayed None), so true acoustic boosting remains unproven work. The intensifier suffix is accepted for Deepgram wire compatibility but does not weight the correction. Gated by DDX_STT_TERM_CORRECTION_ENABLED.
`interim_results`	SUPPORTED	Stream partial Results frames; if false buffer to final only.
`utterance_end_ms`	SUPPORTED	Silence (ms) after last word before UtteranceEnd fires; drives SDK end-of-speech timing. None falls back to VAD default.
`vad_events`	SUPPORTED	Silero/WebRTC VAD; emit SpeechStarted + UtteranceEnd frames.
`words`	SUPPORTED	Parakeet provides word-level timestamps natively.

PARTIAL Partial (11)

Param	Behaviour	Note
`channels`	PARTIAL	Accept up to 2; mix to mono internally.
`model`	PARTIAL	Accept any value, route to Parakeet, echo in metadata.model_info.name.
`punctuate`	PARTIAL	Parakeet basic punctuation; post-processor for accuracy.
`diarize`	PARTIAL	Baseline speaker:0. When DDX_STT_DIARIZE_ENABLED=1, real Sortformer speaker labels on words[] (en supported; de/fr/es EXPERIMENTAL).
`smart_format`	PARTIAL	Baseline: punctuate + numerals. When DDX_STT_SMART_FORMAT_ENABLED=1, engages WFST ITN on the FINAL transcript (en production; de/fr/es EXPERIMENTAL).
`replace`	PARTIAL	Repeatable 'search:replace'; simple string-replace post-processor.
`utterances`	PARTIAL	Per-utterance segmentation via VAD + utterance_end_ms silence.
`utt_split`	PARTIAL	Silence threshold seconds; map to VAD min_silence_ms.
`turn_detection`	PARTIAL	Dudoxx ext (shard 06). Silero endpointing is default. When DDX_STT_TURN_DETECTOR_ENABLED=1, a semantic EOU model REFINES each Silero speech_end (true turn-end vs mid-utterance pause); off = Silero only.
`turn_decision`	PARTIAL	Dudoxx ext. When DDX_STT_TURN_DECISION_ENABLED=1, every is_final Results frame carries ddx_turn {decision: continue\|complete\|incomplete, status, source: llm\|livekit, ...} — a dudoxx-LLM verdict over the turn text since the last complete, LiveKit EOU fallback (2-state). Advisory; never re-times a finalize.
`filler_words`	PARTIAL	Parakeet drops fillers by default; toggling needs model-config change.

NOOP Noop (20)

Param	Behaviour	Note
`version`	NOOP	Legacy Deepgram field; accept silently, ignore.
`tier`	NOOP	Legacy Deepgram field; accept silently, ignore.
`profanity_filter`	NOOP	Accepted and echoed; the transcript is returned unfiltered. No profanity filtering is applied.
`redact`	NOOP	Accepted and echoed; the transcript is returned unredacted. No redaction is applied — do not rely on this param for PII handling.
`keyterm`	NOOP	ACCEPTED BUT INERT: Deepgram's Nova-3 keyterm surface (max 500 tokens) is parsed and echoed, but has no effect on the transcript — same decoder-side gap as 'keywords'. Not implemented, no acoustic or post-processing effect.
`no_delay`	NOOP	DG smart_format no-delay flag; our streaming path emits without an endpointing buffer delay already, so honour-but-noop (echoed).
`tag`	NOOP	Repeatable opaque analytics tag; accept silently, log only.
`extra`	NOOP	Opaque echo-back string; surface in opening Metadata transaction_key.
`mip_opt_out`	NOOP	Model-improvement opt-out; honour silently (no audio retention anyway).
`callback`	NOOP	Async callback URL; not applicable to streaming WS.
`callback_method`	NOOP	POST/GET/PUT/DELETE for async callback; ignored in streaming.
`paragraphs`	NOOP	Accept; omit paragraph structure from response.
`summarize`	NOOP	Invariant 8: NEVER call server-side LLM; return null/empty.
`detect_language`	NOOP	Invariant 8: NEVER call server-side LLM; return null languages.
`detect_topics`	NOOP	Invariant 8: NEVER call server-side LLM; return empty topics.
`detect_entities`	NOOP	Invariant 8: NEVER call server-side LLM; return entities:[].
`dictation`	NOOP	Spoken commands as regular words; no command interpreter.
`search`	NOOP	Repeatable; omit search_results from response.
`alternatives`	NOOP	Always return 1 alternative; SDK handles missing N-best gracefully.
`authorization`	NOOP	SDK per-request key override; the WS handler reads Authorization header.

ERROR Error (1)

Param	Behaviour	Note
`multichannel`	ERROR	Reject if true with channels>1: close 1008 'multichannel not supported'.

Opt-in capabilities (language support)

These params are honoured ONLY when their server env flag is on; otherwise they keep today's baseline behaviour (speaker:0 / verbatim text / Silero-only endpointing). EXPERIMENTAL languages run but correctness is not guaranteed.

Param	Production langs	EXPERIMENTAL langs	Notes
`diarize`	`en`	`de`, `es`, `fr`	Sortformer speaker labels on `words[].speaker` (additive extension field). Gated by `DDX_STT_DIARIZE_ENABLED`.
`smart_format`	`en`	`de`, `es`, `fr`	WFST inverse text normalization on the FINAL transcript. Gated by `DDX_STT_SMART_FORMAT_ENABLED`.
`turn_detection`	`de`, `en`, `es`, `fr`, `it`, `nl`, `pt`	—	Semantic end-of-utterance refinement over Silero (fail-open). Gated by `DDX_STT_TURN_DETECTOR_ENABLED`.
`turn_decision`	LLM: any	LiveKit fallback: its own language set (others → unavailable)	Advisory `ddx_turn` object on every `is_final` Results frame: `decision` = `continue` \| `complete` \| `incomplete` (or null), plus `status` and `source` (`llm` \| `livekit` \| null). Never re-times a finalize. Gated by `DDX_STT_TURN_DECISION_ENABLED`.