@absolutejs/voice 0.0.22-beta.525 → 0.0.22-beta.527

package/README.md

@@ -4229,24 +4229,6 @@ await voice.startRecording();
 - partial user speech starts arriving
 - manual `sendAudio(...)` is called while assistant audio is playing
 
-## Duplex Benchmarks
-
-The first duplex benchmark lane measures package-level barge-in interruption on the client path. It records scenario pass/fail plus local interruption latency for:
-
-- manual `sendAudio(...)`
-- partial transcript start
-- input-level threshold crossing
-
-Run it with:
-
-```bash
-bun run bench:duplex
-```
-
-That writes:
-
-- `benchmark-results/duplex-barge-in.json`
-
 ## Telephony
 
 `@absolutejs/voice` now includes a first PSTN bridge layer for Twilio Media Streams. It converts inbound `audio/x-mulaw` 8 kHz frames into the PCM format the voice session expects, and converts assistant PCM audio back into outbound Twilio media events.
@@ -4293,158 +4275,16 @@ await bridge.handleMessage(mediaMessageFromTwilio);
 
 The bridge also sends Twilio `clear` events on new inbound media after assistant audio has started streaming, so telephony barge-in can stop queued outbound playback.
 
-You can benchmark the package-level Twilio bridge path with:
-
-```bash
-bun run bench:telephony:run
-```
-
-That writes:
-
-- `benchmark-results/telephony-twilio-bridge.json`
-- `benchmark-results/telephony-run-manifest.json`
-
-For a live vendor-backed duplex smoke benchmark on the real TTS adapters, run:
-
-```bash
-bun run bench:duplex:live:run
-```
-
-That writes fresh results to:
-
-For a live vendor-backed telephony smoke benchmark through the Twilio bridge path, run:
-
-```bash
-bun run bench:telephony:live:run
-```
-
-That writes:
-
-- `benchmark-results/telephony-live-deepgram-elevenlabs.json`
-- `benchmark-results/telephony-live-run-manifest.json`
-
-For a repeated live telephony stability read, run:
-
-```bash
-bun run bench:telephony:live:series
-```
-
-That writes:
-
-- `benchmark-results/telephony-live-series-summary-runs-3.json`
-
-For a live Deepgram telephony model shootout on the same PSTN path, run:
-
-```bash
-bun run bench:telephony:live:shootout
-```
-
-That writes:
-
-- `benchmark-results/telephony-live-flux-general-en.json`
-- `benchmark-results/telephony-live-nova-3-phone.json`
-- `benchmark-results/telephony-live-shootout-manifest.json`
-
-- `benchmark-results/duplex-live-elevenlabs.json`
-- `benchmark-results/duplex-live-openai.json`
-- `benchmark-results/duplex-live-all.json`
-- `benchmark-results/duplex-live-run-manifest.json`
-
-For a browser-run duplex benchmark that uses a real headless Chrome `AudioContext` instead of the fake Node-side playback context, run:
-
-```bash
-bun run bench:duplex:browser:run
-```
-
-That writes fresh results to:
-
-- `benchmark-results/duplex-browser-elevenlabs.json`
-- `benchmark-results/duplex-browser-openai.json`
-- `benchmark-results/duplex-browser-all.json`
-- `benchmark-results/duplex-browser-run-manifest.json`
-
-To measure browser duplex stability across repeated runs, use:
-
-```bash
-bun run bench:duplex:browser:series
-```
-
-That writes:
-
-- `benchmark-results/duplex-browser-series-summary-runs-3.json`
-- per-run provider artifacts like `benchmark-results/duplex-browser-elevenlabs-series-run-1.json`
-
-For repeated interrupt-and-resume across several consecutive assistant turns, run:
-
-```bash
-bun run bench:duplex:browser:overlap:run
-```
-
-That writes:
-
-- `benchmark-results/duplex-browser-overlap-elevenlabs.json`
-- `benchmark-results/duplex-browser-overlap-openai.json`
-- `benchmark-results/duplex-browser-overlap-all.json`
-- `benchmark-results/duplex-browser-overlap-run-manifest.json`
-
-To measure overlap stability across repeated live browser runs, use:
-
-```bash
-bun run bench:duplex:browser:overlap:series
-```
-
-That writes:
-
-- `benchmark-results/duplex-browser-overlap-series-summary-runs-3.json`
-- per-run provider artifacts like `benchmark-results/duplex-browser-overlap-elevenlabs-series-run-1.json`
-
-## TTS Benchmarks
-
-`@absolutejs/voice` now includes a first TTS benchmark harness for streaming output adapters. The initial metrics are:
-
-- `firstAudioLatencyMs`
-- `elapsedMs`
-- `audioChunkCount`
-- `totalAudioBytes`
-- estimated PCM `audioDurationMs`
-- interruption responsiveness via `interruptionLatencyMs`
-
-Run the full TTS suite with one command:
-
-```bash
-bun run bench:tts:run
-```
-
-That writes fresh results to:
-
-- `benchmark-results/tts-all.json`
-- `benchmark-results/tts-elevenlabs.json`
-- `benchmark-results/tts-openai.json`
-- `benchmark-results/tts-run-manifest.json`
-
-To measure interruption/cancel responsiveness separately:
-
-```bash
-bun run bench:tts:interrupt:run
-```
-
-That writes fresh interruption results to:
-
-- `benchmark-results/tts-all-interrupt.json`
-- `benchmark-results/tts-elevenlabs-interrupt.json`
-- `benchmark-results/tts-openai-interrupt.json`
-- `benchmark-results/tts-interrupt-run-manifest.json`
-
 ## Recommended Production Path
 
-The current best-performing path in the bundled benchmarks is:
+The current best-performing path is:
 
 - `deepgram-flux` as primary STT
 - route-level `lexicon` for pronunciation/domain entries
 - route-level `phraseHints`
 - route-level `correctTurn` using `createPhraseHintCorrectionHandler()`
 
-That combination outperformed the raw vendor-only paths in the package benchmarks because it lets AbsoluteJS repair domain-specific terms after strong base transcription instead of depending on a second STT vendor to rescue hard turns.
+That combination outperformed the raw vendor-only paths in our benchmarks ([absolutejs/benchmarks](https://github.com/absolutejs/benchmarks)) because it lets AbsoluteJS repair domain-specific terms after strong base transcription instead of depending on a second STT vendor to rescue hard turns.
 
 Minimal production-oriented example:
 
@@ -4527,11 +4367,7 @@ app.use(
 
 - `best` maps to the current strongest in-package path: Deepgram Flux plus generic deterministic correction.
 - `low-cost` maps to a cheaper/raw package path: one primary STT pass with no correction hook.
-- session benchmarks now include per-turn cost telemetry fields like `averageRelativeCostUnits`, `averagePrimaryAudioMs`, and `averageFallbackReplayAudioMs`.
-- use `bun run bench:stt:routing:run` to benchmark both in parallel and write fresh:
-  - `benchmark-results/sessions-best-stt-runs-3.json`
-  - `benchmark-results/sessions-cheap-stt-runs-3.json`
-  - `benchmark-results/stt-routing-run-manifest.json`
+- session cost telemetry exposes per-turn fields like `averageRelativeCostUnits`, `averagePrimaryAudioMs`, and `averageFallbackReplayAudioMs`.
 
 ## LLM Provider Routing
 
@@ -5175,137 +5011,9 @@ voice.bindHTMX({ element: "#voice-htmx-sync" });
 
 That keeps HTMX pages declarative without inventing custom fragment endpoints for core voice session UI.
 
-## Competitive Benchmarking
-
-The package includes a competitive benchmark harness for STT quality and responsiveness.
-
-Run:
-
-```bash
-bun run bench:vs
-```
-
-Use profiles to focus where you want to win:
-
-- `bun run bench:vs all` (default)
-- `bun run bench:vs all accents`
-- `bun run bench:vs all code-switch`
-- `bun run bench:vs all jargon`
-- `bun run bench:vs all multilingual`
-- `bun run bench:vs all multi-speaker`
-- `bun run bench:vs all telephony`
-- `bun run bench:vs all clean`
-- `bun run bench:vs all noisy`
-- `bun run bench:vs deepgram accents`
-- `bun run bench:vs deepgram-flux accents` (compare Flux candidate, default includes VAPI output if configured)
-- `bun run bench:vs deepgram-nova accents`
-
-Current benchmark guidance:
-
-- use `deepgram-flux` as the primary conversational STT path
-- prefer route-level `phraseHints` plus `correctTurn` over cross-vendor fallback for domain-specific accuracy
-- use fallback vendors only when your own traffic proves they beat the package-level correction path
-- do not treat `openai` as the default STT path unless your own benchmarks prove it for your traffic
-
-If you use a VAPI baseline file, you can run a direct model comparison:
-
-```bash
-bun run bench:vs:deepgram-flux
-```
-
-To benchmark Nova vs Flux back-to-back, set the model explicitly:
-
-```bash
-DEEPGRAM_MODEL=flux-general-en bun run bench:deepgram:accents
-DEEPGRAM_MODEL=nova-3 bun run bench:deepgram:accents
-```
-
-To stress the STT path with synthesized narrowband phone audio:
-
-```bash
-bun run bench:telephony
-bun run bench:telephony:run
-bun run bench:deepgram:telephony
-bun run bench:deepgram:corrected:telephony
-bun run bench:jargon
-bun run bench:deepgram:jargon
-bun run bench:deepgram:corrected:audit:jargon
-bun run bench:multi-speaker:run
-bun run bench:multi-speaker:analyze
-bun run bench:deepgram:multi-speaker
-```
-
-To compare against Vapi or other providers, provide a baseline JSON file:
-
-```bash
-bun run bench:vs all accents --compare /path/to/vapi-baseline.json
-```
-
-Expected benchmark payload:
-
-```json
-{
-  "source": "vapi",
-  "results": [
-    {
-      "adapterId": "vapi-baseline",
-      "summary": {
-        "passRate": 0.0,
-        "averageWordErrorRate": 1.0,
-        "averageTermRecall": 0.0,
-        "averageElapsedMs": 0,
-        "averageTimeToEndOfTurnMs": 0,
-        "averageTimeToFirstFinalMs": 0,
-        "averageTimeToFirstPartialMs": 0,
-        "wordAccuracyRate": 0.0
-      }
-    }
-  ]
-}
-```
-
-For a fast parse-only validation of arguments:
-
-```bash
-bun run ./scripts/benchmark-vs.ts --dry-run
-```
-
-The harness prints:
-
-- pass rate and recall deltas per adapter
-- weighted scorecard (`passRate`, term recall, word accuracy)
-- optional competitor deltas (Vapi)
-- a markdown report beside the JSON output, for example:
-  - `benchmark-results/vs-all-telephony.json`
-  - `benchmark-results/vs-all-telephony.md`
-
-For package-level multi-turn behavior, use the session benchmark harness instead of raw STT-only benchmarking:
-
-```bash
-bun run bench:sessions
-bun run bench:deepgram:sessions
-bun run bench:deepgram:soak:sessions
-bun run bench:deepgram:hybrid:sessions
-bun run bench:deepgram:corrected:sessions
-bun run bench:deepgram:corrected:soak:sessions
-bun run bench:stt:routing:run
-bun run bench:assemblyai:sessions
-bun run bench:openai:sessions
-bun run bench:soak:run
-```
-
-That harness runs the adapter through `VoiceSession` itself, so the output reflects reconnect handling, turn commit stability, and duplicate-turn protection rather than only raw transcript quality.
-
-`bench:soak:run` is the STT-5 runner. It executes the long-session soak lane for raw Deepgram Flux, corrected Deepgram, and the reconnect resilience suite in parallel, then writes fresh JSON into `benchmark-results/` without the runs deleting each other.
-
-`bench:stt:routing:run` is the STT-7 runner. It benchmarks the package’s current `best` vs `low-cost` session strategies in parallel, clears stale outputs first, and writes a manifest so the cost-aware summaries are guaranteed fresh.
+## Benchmarks
 
-`bench:deepgram:corrected:sessions` exercises the current recommended package-level production path:
-
-- Deepgram Flux as primary STT
-- phrase hints routed through the adapter layer
-- committed-turn correction via `createPhraseHintCorrectionHandler()`
-- core turn dedupe, reconnect, and transcript selection still owned by `@absolutejs/voice`
+Performance & accuracy benchmarks (STT, TTS, duplex, telephony, sessions) and head-to-head comparisons against Vapi live in a dedicated repo: **[absolutejs/benchmarks](https://github.com/absolutejs/benchmarks)**. They consume the published `@absolutejs/voice` package and provider adapters.
 
 ## Adapter Contract
 
@@ -5426,100 +5134,6 @@ Fallback triggers are evaluated at commit time:
 
 The fallback adapter receives the same window of turn audio as the primary (default `8s`, configurable with `replayWindowMs`) and can only run `maxAttemptsPerTurn` times per turn.
 
-## Benchmark Fixture Sources
-
-Bundled fixtures cover the current in-repo English benchmark suite. For multilingual and code-switch evaluation, add external fixture directories and let the benchmark scripts merge them automatically.
-
-The public corpus builder currently assembles:
-
-- FLEURS multilingual dev clips
-- BSC Catalan-Spanish code-switch evaluation clips
-- CoSHE Hindi-English code-switch evaluation clips
-
-Set either:
-
-- `VOICE_FIXTURE_DIR=/abs/path/to/fixtures`
-- `VOICE_FIXTURE_DIRS=/abs/path/one,/abs/path/two`
-
-Each fixture directory must include:
-
-- `manifest.json`
-- `pcm/*.pcm`
-
-Each manifest entry can include:
-
-- `language`
-- `tags`
-  Use `multilingual`, `bilingual`, or `code-switch` to route fixtures into the multilingual benchmark lane.
-
-Benchmark commands:
-
-```bash
-bun run bench:multilingual
-bun run bench:code-switch
-bun run bench:code-switch:series
-bun run bench:code-switch:ca-es
-bun run bench:code-switch:ca-es:series
-bun run bench:code-switch:ca-es:corts:series
-bun run bench:code-switch:ca-es:parlament:series
-bun run bench:code-switch:hi-en
-bun run bench:code-switch:hi-en:series
-bun run bench:deepgram:multilingual
-bun run bench:deepgram:code-switch
-bun run bench:deepgram:code-switch:series
-bun run bench:deepgram:code-switch:ca-es
-bun run bench:deepgram:code-switch:ca-es:series
-bun run bench:deepgram:code-switch:ca-es:corts:series
-bun run bench:deepgram:code-switch:ca-es:parlament:series
-bun run bench:deepgram:code-switch:ca-es:nova3-multi:series
-bun run bench:deepgram:code-switch:ca-es:nova3-ca:series
-bun run bench:deepgram:code-switch:ca-es:nova3-es:series
-bun run bench:deepgram:code-switch:ca-es:nova2-ca:series
-bun run bench:deepgram:code-switch:ca-es:nova2-es:series
-bun run bench:deepgram:code-switch:ca-es:best:corrected:series
-bun run bench:deepgram:code-switch:ca-es:parlament:debug
-bun run bench:deepgram:code-switch:corrected:ca-es
-bun run bench:deepgram:code-switch:corrected:ca-es:series
-bun run bench:deepgram:code-switch:corrected:ca-es:corts:series
-bun run bench:deepgram:code-switch:corrected:ca-es:parlament:series
-bun run bench:deepgram:code-switch:hi-en
-bun run bench:deepgram:code-switch:hi-en:series
-bun run bench:deepgram:code-switch:corrected:hi-en
-bun run bench:deepgram:code-switch:corrected:hi-en:series
-bun run bench:deepgram:code-switch:corrected
-bun run bench:deepgram:code-switch:corrected:series
-bun run bench:assemblyai:multilingual
-bun run bench:assemblyai:code-switch
-bun run bench:openai:multilingual
-bun run bench:openai:code-switch
-bun run bench:openai:code-switch:series
-bun run bench:openai:code-switch:ca-es
-bun run bench:openai:code-switch:ca-es:series
-bun run bench:openai:code-switch:corrected:ca-es
-bun run bench:openai:code-switch:corrected:ca-es:series
-bun run bench:openai:code-switch:hi-en
-bun run bench:openai:code-switch:hi-en:series
-bun run bench:openai:code-switch:corrected:hi-en
-bun run bench:openai:code-switch:corrected:hi-en:series
-bun run bench:openai:code-switch:corrected
-bun run bench:openai:code-switch:corrected:series
-```
-
-Current benchmark direction:
-
-- `openai` is the strongest adapter on the current public multilingual corpus
-- `deepgram` remains the strongest browser-English path
-- raw code-switch remains a weaker surface for every adapter and should be benchmarked separately with `bench:code-switch`
-- jargon-heavy/domain-heavy English terms now have their own profile; use `bench:jargon` for the cross-adapter read and `bench:deepgram:corrected:audit:jargon` to compare `raw` vs `generic` vs `experimental` vs `benchmarkSeeded`
-- code-switch should be treated as language-pair-specific, not one universal lane; `ca-es` and `hi-en` now have dedicated series commands
-- `ca-es` also has a dedicated Deepgram model/language shootout lane so you can compare `nova-3`/`nova-2` with `multi`, `ca`, and `es` routing without overwriting results
-- current best `ca-es` base path is `deepgram` `nova-3` with `language=ca`; the short runner script uses that path for corrected series
-- `ca-es` is also split by source now: `corts_valencianes` and `parlament_parla` can be benchmarked independently, and `parlament_parla` has a dedicated transcript dump script
-- corrected code-switch runs now have dedicated lexicon-driven series commands so raw and corrected stability can be compared directly
-- multi-speaker diarization is now its own benchmark surface; use `bench:multi-speaker:run` for the parallel cross-adapter plus Deepgram-specific read
-- when tuning diarization specifically, use `bench:multi-speaker:analyze` to split Deepgram into clean vs noisy handoff lanes, include a corrected noisy read, and emit a speaker-pattern debug dump
-- use the `:series` commands when you need stability rather than a single-pass snapshot
-
 ## Client Primitives
 
 Browser and framework helpers sit on top of the same connection core:

package/dist/index.d.ts

@@ -204,7 +204,7 @@ export { buildVoicePostCallAnalysisReport, createVoicePostCallAnalysisRoutes, re
 export { buildVoiceGuardrailReport, createVoiceGuardrailPolicy, createVoiceGuardrailRuntime, createVoiceGuardrailRoutes, evaluateVoiceGuardrailPolicy, renderVoiceGuardrailMarkdown, voiceGuardrailPolicyPresets, } from "./guardrails";
 export { createId, createVoiceSessionRecord } from "./store";
 export { createVoiceSTTRoutingCorrectionHandler, resolveVoiceSTTRoutingStrategy, } from "./routing";
-export { applyRiskTieredPhraseHintCorrections, applyPhraseHintCorrections, createDomainLexicon, createDomainPhraseHints, createPhraseHintCorrectionHandler, createRiskyTurnCorrectionHandler, } from "./correction";
+export { applyLexiconCorrections, applyRiskTieredPhraseHintCorrections, applyPhraseHintCorrections, createDomainLexicon, createDomainPhraseHints, createPhraseHintCorrectionHandler, createRiskyTurnCorrectionHandler, } from "./correction";
 export { conditionAudioChunk, resolveAudioConditioningConfig, } from "./audioConditioning";
 export { resolveVoiceRuntimePreset } from "./presets";
 export { resolveTurnDetectionConfig, TURN_PROFILE_DEFAULTS, } from "./turnProfiles";

package/dist/index.js

@@ -53040,6 +53040,7 @@ export {
   applyVoiceCampaignTelephonyOutcome,
   applyRiskTieredPhraseHintCorrections,
   applyPhraseHintCorrections,
+  applyLexiconCorrections,
   applyBrowserNoiseSuppression,
   appendVoiceRealCallProfileRecoveryEvidence,
   appendVoiceProviderRouterTraceEvent,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@absolutejs/voice",
-  "version": "0.0.22-beta.525",
+  "version": "0.0.22-beta.527",
   "description": "Voice primitives and Elysia plugin for AbsoluteJS",
   "repository": {
     "type": "git",
@@ -16,150 +16,6 @@
   "license": "CC BY-NC 4.0",
   "author": "Alex Kahn",
   "scripts": {
-    "bench:accents": "bun run ./scripts/benchmark-stt.ts all accents",
-    "bench:jargon": "bun run ./scripts/benchmark-stt.ts all jargon",
-    "bench:jargon:run": "bun run ./scripts/run-jargon-benchmarks.ts",
-    "bench:multilingual": "bun run ./scripts/benchmark-stt.ts all multilingual",
-    "bench:multi-speaker": "bun run ./scripts/benchmark-stt.ts all multi-speaker",
-    "bench:multi-speaker:run": "bun run ./scripts/run-multi-speaker-benchmarks.ts",
-    "bench:multi-speaker:analyze": "bun run ./scripts/run-multi-speaker-analysis.ts",
-    "bench:deepgram:multi-speaker:clean": "bun run ./scripts/benchmark-stt.ts deepgram multi-speaker-clean",
-    "bench:deepgram:multi-speaker:noisy": "bun run ./scripts/benchmark-stt.ts deepgram multi-speaker-noisy",
-    "bench:deepgram:multi-speaker:noisy:corrected": "bun run ./scripts/benchmark-stt.ts deepgram-corrected multi-speaker-noisy",
-    "bench:deepgram:multi-speaker:debug": "bun run ./scripts/run-multi-speaker-debug.ts",
-    "bench:code-switch": "bun run ./scripts/benchmark-stt.ts all code-switch",
-    "bench:code-switch:series": "bun run ./scripts/benchmark-stt.ts all code-switch --runs 5",
-    "bench:code-switch:ca-es": "bun run ./scripts/benchmark-stt.ts all code-switch-ca-es",
-    "bench:code-switch:ca-es:series": "bun run ./scripts/benchmark-stt.ts all code-switch-ca-es --runs 5",
-    "bench:code-switch:ca-es:corts:series": "bun run ./scripts/benchmark-stt.ts all code-switch-ca-es-corts --runs 5",
-    "bench:code-switch:ca-es:parlament:series": "bun run ./scripts/benchmark-stt.ts all code-switch-ca-es-parlament --runs 5",
-    "bench:code-switch:hi-en": "bun run ./scripts/benchmark-stt.ts all code-switch-hi-en",
-    "bench:code-switch:hi-en:series": "bun run ./scripts/benchmark-stt.ts all code-switch-hi-en --runs 5",
-    "bench:telephony": "bun run ./scripts/benchmark-stt.ts all telephony",
-    "bench:telephony:bridge": "bun run ./scripts/benchmark-telephony.ts",
-    "bench:telephony:run": "bun run ./scripts/run-telephony-benchmarks.ts",
-    "bench:tts": "bun run ./scripts/benchmark-tts.ts all",
-    "bench:tts:elevenlabs": "bun run ./scripts/benchmark-tts.ts elevenlabs",
-    "bench:tts:openai": "bun run ./scripts/benchmark-tts.ts openai",
-    "bench:tts:interrupt": "bun run ./scripts/benchmark-tts.ts all interrupt",
-    "bench:tts:interrupt:elevenlabs": "bun run ./scripts/benchmark-tts.ts elevenlabs interrupt",
-    "bench:tts:interrupt:openai": "bun run ./scripts/benchmark-tts.ts openai interrupt",
-    "bench:tts:run": "bun run ./scripts/run-tts-benchmarks.ts",
-    "bench:tts:interrupt:run": "bun run ./scripts/run-tts-interruption-benchmarks.ts",
-    "bench:duplex": "bun run ./scripts/benchmark-duplex.ts",
-    "bench:duplex:live": "bun run ./scripts/benchmark-live-duplex.ts all",
-    "bench:duplex:live:elevenlabs": "bun run ./scripts/benchmark-live-duplex.ts elevenlabs",
-    "bench:duplex:live:openai": "bun run ./scripts/benchmark-live-duplex.ts openai",
-    "bench:duplex:live:run": "bun run ./scripts/run-live-duplex-benchmarks.ts",
-    "bench:telephony:live": "bun run ./scripts/benchmark-live-telephony.ts",
-    "bench:telephony:live:run": "bun run ./scripts/run-live-telephony-benchmarks.ts",
-    "bench:telephony:live:series": "bun run ./scripts/run-live-telephony-series.ts",
-    "bench:telephony:live:shootout": "bun run ./scripts/run-live-telephony-shootout.ts",
-    "bench:telephony:live:stt:shootout": "bun run ./scripts/run-live-telephony-stt-shootout.ts",
-    "bench:telephony:live:tts:shootout": "bun run ./scripts/run-live-telephony-tts-shootout.ts",
-    "bench:duplex:browser": "bun run ./scripts/benchmark-browser-duplex.ts all",
-    "bench:duplex:browser:elevenlabs": "bun run ./scripts/benchmark-browser-duplex.ts elevenlabs",
-    "bench:duplex:browser:openai": "bun run ./scripts/benchmark-browser-duplex.ts openai",
-    "bench:duplex:browser:run": "bun run ./scripts/run-browser-duplex-benchmarks.ts",
-    "bench:duplex:browser:series": "bun run ./scripts/run-browser-duplex-series.ts",
-    "bench:duplex:browser:overlap": "bun run ./scripts/benchmark-browser-duplex-overlap.ts all",
-    "bench:duplex:browser:overlap:run": "bun run ./scripts/run-browser-duplex-overlap-benchmarks.ts",
-    "bench:duplex:browser:overlap:series": "bun run ./scripts/run-browser-duplex-overlap-series.ts",
-    "bench:production": "bun run ./scripts/benchmark-production.ts",
-    "bench:production:deepgram-corrected:audit": "bun run ./scripts/benchmark-production.ts deepgram-corrected-audit",
-    "bench:assemblyai": "bun run ./scripts/benchmark-stt.ts assemblyai",
-    "bench:assemblyai:accents": "bun run ./scripts/benchmark-stt.ts assemblyai accents",
-    "bench:assemblyai:multilingual": "bun run ./scripts/benchmark-stt.ts assemblyai multilingual",
-    "bench:assemblyai:code-switch": "bun run ./scripts/benchmark-stt.ts assemblyai code-switch",
-    "bench:openai": "bun run ./scripts/benchmark-stt.ts openai",
-    "bench:openai:accents": "bun run ./scripts/benchmark-stt.ts openai accents",
-    "bench:openai:multilingual": "bun run ./scripts/benchmark-stt.ts openai multilingual",
-    "bench:openai:code-switch": "bun run ./scripts/benchmark-stt.ts openai code-switch",
-    "bench:openai:code-switch:series": "bun run ./scripts/benchmark-stt.ts openai code-switch --runs 5",
-    "bench:openai:code-switch:ca-es": "bun run ./scripts/benchmark-stt.ts openai code-switch-ca-es",
-    "bench:openai:code-switch:ca-es:series": "bun run ./scripts/benchmark-stt.ts openai code-switch-ca-es --runs 5",
-    "bench:openai:code-switch:hi-en": "bun run ./scripts/benchmark-stt.ts openai code-switch-hi-en",
-    "bench:openai:code-switch:hi-en:series": "bun run ./scripts/benchmark-stt.ts openai code-switch-hi-en --runs 5",
-    "bench:openai:code-switch:corrected": "bun run ./scripts/benchmark-stt.ts openai-corrected code-switch",
-    "bench:openai:code-switch:corrected:series": "bun run ./scripts/benchmark-stt.ts openai-corrected code-switch --runs 5",
-    "bench:openai:code-switch:corrected:ca-es": "bun run ./scripts/benchmark-stt.ts openai-corrected code-switch-ca-es",
-    "bench:openai:code-switch:corrected:ca-es:series": "bun run ./scripts/benchmark-stt.ts openai-corrected code-switch-ca-es --runs 5",
-    "bench:openai:code-switch:corrected:hi-en": "bun run ./scripts/benchmark-stt.ts openai-corrected code-switch-hi-en",
-    "bench:openai:code-switch:corrected:hi-en:series": "bun run ./scripts/benchmark-stt.ts openai-corrected code-switch-hi-en --runs 5",
-    "bench:openai:diag": "bun run ./scripts/benchmark-stt.ts openai --diag",
-    "bench:deepgram": "bun run ./scripts/benchmark-stt.ts deepgram",
-    "bench:deepgram:accents": "bun run ./scripts/benchmark-stt.ts deepgram accents",
-    "bench:deepgram:jargon": "bun run ./scripts/benchmark-stt.ts deepgram jargon",
-    "bench:deepgram:multilingual": "bun run ./scripts/benchmark-stt.ts deepgram multilingual",
-    "bench:deepgram:multi-speaker": "bun run ./scripts/benchmark-stt.ts deepgram multi-speaker",
-    "bench:deepgram:code-switch": "bun run ./scripts/benchmark-stt.ts deepgram code-switch",
-    "bench:deepgram:code-switch:series": "bun run ./scripts/benchmark-stt.ts deepgram code-switch --runs 5",
-    "bench:deepgram:code-switch:ca-es": "bun run ./scripts/benchmark-stt.ts deepgram code-switch-ca-es",
-    "bench:deepgram:code-switch:ca-es:series": "bun run ./scripts/benchmark-stt.ts deepgram code-switch-ca-es --runs 5",
-    "bench:deepgram:code-switch:ca-es:corts:series": "bun run ./scripts/benchmark-stt.ts deepgram code-switch-ca-es-corts --runs 5",
-    "bench:deepgram:code-switch:ca-es:parlament:series": "bun run ./scripts/benchmark-stt.ts deepgram code-switch-ca-es-parlament --runs 5",
-    "bench:deepgram:code-switch:ca-es:nova3-multi:series": "DEEPGRAM_CODE_SWITCH_MODEL=nova-3 DEEPGRAM_CODE_SWITCH_LANGUAGE=multi bun run ./scripts/benchmark-stt.ts deepgram code-switch-ca-es --runs 5 --variant nova3-multi",
-    "bench:deepgram:code-switch:ca-es:nova3-ca:series": "DEEPGRAM_CODE_SWITCH_MODEL=nova-3 DEEPGRAM_CODE_SWITCH_LANGUAGE=ca bun run ./scripts/benchmark-stt.ts deepgram code-switch-ca-es --runs 5 --variant nova3-ca",
-    "bench:deepgram:code-switch:ca-es:nova3-es:series": "DEEPGRAM_CODE_SWITCH_MODEL=nova-3 DEEPGRAM_CODE_SWITCH_LANGUAGE=es bun run ./scripts/benchmark-stt.ts deepgram code-switch-ca-es --runs 5 --variant nova3-es",
-    "bench:deepgram:code-switch:ca-es:nova2-ca:series": "DEEPGRAM_CODE_SWITCH_MODEL=nova-2 DEEPGRAM_CODE_SWITCH_LANGUAGE=ca bun run ./scripts/benchmark-stt.ts deepgram code-switch-ca-es --runs 5 --variant nova2-ca",
-    "bench:deepgram:code-switch:ca-es:nova2-es:series": "DEEPGRAM_CODE_SWITCH_MODEL=nova-2 DEEPGRAM_CODE_SWITCH_LANGUAGE=es bun run ./scripts/benchmark-stt.ts deepgram code-switch-ca-es --runs 5 --variant nova2-es",
-    "bench:deepgram:code-switch:ca-es:best:corrected:series": "bun run ./scripts/run-caes-corrected-benchmark.ts",
-    "bench:deepgram:code-switch:hi-en": "bun run ./scripts/benchmark-stt.ts deepgram code-switch-hi-en",
-    "bench:deepgram:code-switch:hi-en:series": "bun run ./scripts/benchmark-stt.ts deepgram code-switch-hi-en --runs 5",
-    "bench:deepgram:code-switch:corrected": "bun run ./scripts/benchmark-stt.ts deepgram-corrected code-switch",
-    "bench:deepgram:code-switch:corrected:series": "bun run ./scripts/benchmark-stt.ts deepgram-corrected code-switch --runs 5",
-    "bench:deepgram:code-switch:corrected:ca-es": "bun run ./scripts/benchmark-stt.ts deepgram-corrected code-switch-ca-es",
-    "bench:deepgram:code-switch:corrected:ca-es:series": "bun run ./scripts/benchmark-stt.ts deepgram-corrected code-switch-ca-es --runs 5",
-    "bench:deepgram:code-switch:corrected:ca-es:corts:series": "bun run ./scripts/benchmark-stt.ts deepgram-corrected code-switch-ca-es-corts --runs 5",
-    "bench:deepgram:code-switch:corrected:ca-es:parlament:series": "bun run ./scripts/benchmark-stt.ts deepgram-corrected code-switch-ca-es-parlament --runs 5",
-    "bench:deepgram:code-switch:ca-es:parlament:debug": "bun run ./scripts/run-caes-parlament-debug.ts",
-    "bench:deepgram:code-switch:ca-es:parlament:rerun": "bun run ./scripts/run-caes-parlament-series.ts",
-    "bench:deepgram:code-switch:ca-es:full:rerun": "bun run ./scripts/run-caes-full-series.ts",
-    "bench:deepgram:code-switch:corrected:hi-en": "bun run ./scripts/benchmark-stt.ts deepgram-corrected code-switch-hi-en",
-    "bench:deepgram:code-switch:corrected:hi-en:series": "bun run ./scripts/benchmark-stt.ts deepgram-corrected code-switch-hi-en --runs 5",
-    "corpus:multilingual": "bun run ./scripts/build-multilingual-corpus.ts",
-    "bench:deepgram:corrected:audit": "DEEPGRAM_MODEL=flux-general-en bun run ./scripts/benchmark-stt.ts deepgram-corrected-audit all",
-    "bench:deepgram:corrected:audit:jargon": "DEEPGRAM_MODEL=flux-general-en bun run ./scripts/benchmark-stt.ts deepgram-corrected-audit jargon",
-    "bench:deepgram:corrected:telephony": "DEEPGRAM_MODEL=flux-general-en bun run ./scripts/benchmark-stt.ts deepgram-corrected telephony",
-    "bench:deepgram:corrected:audit:telephony": "DEEPGRAM_MODEL=flux-general-en bun run ./scripts/benchmark-stt.ts deepgram-corrected-audit telephony",
-    "bench:deepgram:telephony": "bun run ./scripts/benchmark-stt.ts deepgram telephony",
-    "bench:deepgram:nova": "bun run ./scripts/benchmark-stt.ts deepgram",
-    "bench:deepgram:flux": "DEEPGRAM_MODEL=flux-general-en bun run ./scripts/benchmark-stt.ts deepgram",
-    "bench:vs": "bun run ./scripts/benchmark-vs.ts",
-    "bench:vs:all": "bun run ./scripts/benchmark-vs.ts all all",
-    "bench:vs:all:accents": "bun run ./scripts/benchmark-vs.ts all accents",
-    "bench:vs:all:code-switch": "bun run ./scripts/benchmark-vs.ts all code-switch",
-    "bench:vs:all:jargon": "bun run ./scripts/benchmark-vs.ts all jargon",
-    "bench:vs:all:multilingual": "bun run ./scripts/benchmark-vs.ts all multilingual",
-    "bench:vs:all:multi-speaker": "bun run ./scripts/benchmark-vs.ts all multi-speaker",
-    "bench:vs:all:telephony": "bun run ./scripts/benchmark-vs.ts all telephony",
-    "bench:vs:deepgram": "bun run ./scripts/benchmark-vs.ts deepgram all",
-    "bench:vs:deepgram-nova": "bun run ./scripts/benchmark-vs.ts deepgram-nova all",
-    "bench:vs:deepgram-flux": "bun run ./scripts/benchmark-vs.ts deepgram-flux all --compare benchmarks/your-vapi-metrics.json",
-    "bench:vs:assemblyai": "bun run ./scripts/benchmark-vs.ts assemblyai all",
-    "bench:vs:openai": "bun run ./scripts/benchmark-vs.ts openai all",
-    "bench:deepgram:sessions": "bun run ./scripts/benchmark-session.ts deepgram",
-    "bench:stt:best:sessions": "bun run ./scripts/benchmark-session.ts best-stt",
-    "bench:stt:cheap:sessions": "bun run ./scripts/benchmark-session.ts cheap-stt",
-    "bench:stt:routing:run": "bun run ./scripts/run-stt-routing-benchmarks.ts",
-    "bench:deepgram:soak:sessions": "bun run ./scripts/benchmark-session.ts deepgram-flux --profile soak --runs 3",
-    "bench:deepgram:hybrid:sessions": "bun run ./scripts/benchmark-session.ts deepgram-hybrid",
-    "bench:deepgram:corrected:sessions": "bun run ./scripts/benchmark-session.ts deepgram-corrected",
-    "bench:deepgram:corrected:soak:sessions": "bun run ./scripts/benchmark-session.ts deepgram-corrected --profile soak --runs 3",
-    "bench:deepgram:corrected:sessions:audit": "bun run ./scripts/benchmark-session.ts deepgram-corrected-audit",
-    "bench:deepgram:openai-hybrid:sessions": "bun run ./scripts/benchmark-session.ts deepgram-openai-hybrid",
-    "bench:production:deepgram-flux-noisy-room": "bun run ./scripts/benchmark-production.ts deepgram-flux-noisy-room",
-    "bench:production:deepgram-corrected": "bun run ./scripts/benchmark-production.ts deepgram-corrected",
-    "bench:resilience": "bun run ./scripts/benchmark-resilience.ts",
-    "bench:soak:run": "bun run ./scripts/run-session-soak-benchmarks.ts",
-    "bench:soak:debug": "bun run ./scripts/run-session-soak-debug.ts",
-    "bench:soak:debug:corrected": "bun run ./scripts/run-session-soak-debug.ts corrected",
-    "bench:soak:series:debug": "bun run ./scripts/run-session-soak-series-debug.ts raw",
-    "bench:soak:series:debug:corrected": "bun run ./scripts/run-session-soak-series-debug.ts corrected",
-    "bench:sessions": "bun run ./scripts/benchmark-session.ts all",
-    "bench:stt": "bun run ./scripts/benchmark-stt.ts all",
-    "bench:assemblyai:sessions": "bun run ./scripts/benchmark-session.ts assemblyai",
-    "bench:openai:sessions": "bun run ./scripts/benchmark-session.ts openai",
     "build": "bun run ./scripts/build-htmx-bootstrap-asset.ts && rm -rf dist && bun build ./src/index.ts ./src/client/index.ts ./src/react/index.ts ./src/vue/index.ts ./src/svelte/index.ts ./src/angular/index.ts ./src/testing/index.ts --outdir dist --target bun --external elysia --external react --external vue --external @angular/core --external @absolutejs/absolute --external @absolutejs/ai --external @absolutejs/media && bun build ./src/client/htmxBootstrap.ts --outdir dist/client --target browser --format esm && bun build ./src/embed/index.ts --outfile dist/embed/voice-widget.js --target browser --format iife --minify && bun build ./src/embed/index.ts --outdir dist/embed --target browser --format esm && tsc --emitDeclarationOnly --project tsconfig.json",
     "format": "prettier --write \"./**/*.{js,jsx,ts,tsx,json,md}\"",
     "lint": "eslint ./src",