docs: add keywords param to Lightning ASR WS API

fern/apis/waves/asyncapi/lightning-asr-ws.yaml

@@ -158,6 +158,9 @@ servers:
               description:
                 Enable speaker diarization to identify different speakers
                 in the audio.
+            keywords:
+              type: string
+              description: "Comma-separated list of words/phrases to boost, each optionally followed by :INTENSIFIER (e.g. NVIDIA:5,Jensen). Intensifier defaults to 1.0 if omitted. Max 100 keywords per session."
 channels:
   /waves/v1/lightning/get_text:
     address: /waves/v1/lightning/get_text

fern/products/waves/pages/v4.0.0/api-references/lightning-asr-ws.mdx

@@ -34,6 +34,7 @@ The WebSocket connection accepts the following query parameters:
 | `format`              | string | `true`  | Enable punctuation and capitalization formatting in transcript responses. When `true`, transcripts include proper casing and punctuation. When `false`, returns raw lowercase text without punctuation. Options: `true`, `false` |
 | `eou_timeout_ms`      | number | `800`   | End-of-utterance timeout in milliseconds. Controls how long the model waits after the user stops speaking before flushing the transcript as final. Range: `100` to `10000` |
 | `diarize`             | string | `false` | Enable speaker diarization to identify and label different speakers in the audio. When enabled, each word in the transcription includes `speaker` (integer ID) and `speaker_confidence` (float 0-1) fields. Options: `true`, `false` |
+| `keywords`            | string | —       | Comma-separated list of words/phrases to boost, each optionally followed by `:INTENSIFIER` (e.g. `NVIDIA:5,Jensen`). Intensifier defaults to `1.0` if omitted. Max 100 keywords per session. See [Keyword Boosting](/waves/documentation/speech-to-text-pulse/features/keyword-boosting) |
 
 ## Connection Flow
 

fern/products/waves/versions/v4.0.0/api-references/lightning-asr-ws.mdx

@@ -34,6 +34,7 @@ The WebSocket connection accepts the following query parameters:
 | `format`              | string | `true`  | Enable punctuation and capitalization formatting in transcript responses. When `true`, transcripts include proper casing and punctuation. When `false`, returns raw lowercase text without punctuation. Options: `true`, `false` |
 | `eou_timeout_ms`      | number | `800`   | End-of-utterance timeout in milliseconds. Controls how long the model waits after the user stops speaking before flushing the transcript as final. Range: `100` to `10000` |
 | `diarize`             | string | `false` | Enable speaker diarization to identify and label different speakers in the audio. When enabled, each word in the transcription includes `speaker` (integer ID) and `speaker_confidence` (float 0-1) fields. Options: `true`, `false` |
+| `keywords`            | string | —       | Comma-separated list of words/phrases to boost, each optionally followed by `:INTENSIFIER` (e.g. `NVIDIA:5,Jensen`). Intensifier defaults to `1.0` if omitted. Max 100 keywords per session. See [Keyword Boosting](/waves/documentation/speech-to-text-pulse/features/keyword-boosting) |
 
 ## Connection Flow