From 77f0fc85a3fdd9cdda5251f9dfbc0e8118c02656 Mon Sep 17 00:00:00 2001
From: Hoan Luu Huu <110280845+xquanluu@users.noreply.github.com>
Date: Wed, 9 Apr 2025 23:32:06 +0700
Subject: [PATCH] ell tts support speech and pronunciation_dictionary_locators
 (#1137)

---
 lib/tasks/tts-task.js | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/lib/tasks/tts-task.js b/lib/tasks/tts-task.js
index f7f3c43a..a3f31285 100644
--- a/lib/tasks/tts-task.js
+++ b/lib/tasks/tts-task.js
@@ -65,7 +65,6 @@ class TtsTask extends Task {
 
   async setTtsStreamingChannelVars(vendor, language, voice, credentials, ep) {
     const {api_key, model_id, custom_tts_streaming_url, auth_token} = credentials;
-    const {stability, similarity_boost, use_speaker_boost, style} = this.options;
     let obj;
 
     this.logger.debug({credentials},
@@ -87,6 +86,7 @@ class TtsTask extends Task {
         };
         break;
       case 'elevenlabs':
+        const {stability, similarity_boost, use_speaker_boost, style, speed} = this.options.voice_settings || {};
         obj = {
           ELEVENLABS_API_KEY: api_key,
           ELEVENLABS_TTS_STREAMING_MODEL_ID: model_id,
@@ -96,7 +96,14 @@ class TtsTask extends Task {
           ...(stability && {ELEVENLABS_TTS_STREAMING_VOICE_SETTINGS_STABILITY: stability}),
           ...(similarity_boost && {ELEVENLABS_TTS_STREAMING_VOICE_SETTINGS_SIMILARITY_BOOST: similarity_boost}),
           ...(use_speaker_boost && {ELEVENLABS_TTS_STREAMING_VOICE_SETTINGS_USE_SPEAKER_BOOST: use_speaker_boost}),
-          ...(style && {ELEVENLABS_TTS_STREAMING_VOICE_SETTINGS_STYLE: style})
+          ...(style && {ELEVENLABS_TTS_STREAMING_VOICE_SETTINGS_STYLE: style}),
+          // speed has value 0.7 to 1.2, 1.0 is default, make sure we send the value event it's 0
+          ...(speed !== null && speed !== undefined && {ELEVENLABS_TTS_STREAMING_VOICE_SETTINGS_SPEED: `${speed}`}),
+          ...(this.options.pronunciation_dictionary_locators &&
+            Array.isArray(this.options.pronunciation_dictionary_locators) && {
+            ELEVENLABS_TTS_STREAMING_PRONUNCIATION_DICTIONARY_LOCATORS:
+            JSON.stringify(this.options.pronunciation_dictionary_locators)
+          }),
         };
         break;
       case 'rimelabs':