From 043cb1d0853e1dbe835efba4e6a44aa20cb3892d Mon Sep 17 00:00:00 2001
From: Timothy Carambat <rambat1010@gmail.com>
Date: Thu, 17 Jul 2025 10:28:51 -0700
Subject: [PATCH] Add `model` field parameter for generic OpenAI TTS (#4171)

resolves #4165
---
 docker/.env.example                           |  1 +
 .../OpenAiGenericOptions/index.jsx            | 27 ++++++++++++++++---
 server/.env.example                           |  1 +
 server/models/systemSettings.js               |  1 +
 .../utils/TextToSpeech/openAiGeneric/index.js | 10 ++++++-
 server/utils/helpers/updateENV.js             |  4 +++
 6 files changed, 40 insertions(+), 4 deletions(-)

diff --git a/docker/.env.example b/docker/.env.example
index 9c65405c2..d0b6dbeb1 100644
--- a/docker/.env.example
+++ b/docker/.env.example
@@ -255,6 +255,7 @@ GID='1000'
 
 # TTS_PROVIDER="generic-openai"
 # TTS_OPEN_AI_COMPATIBLE_KEY=sk-example
+# TTS_OPEN_AI_COMPATIBLE_MODEL=tts-1
 # TTS_OPEN_AI_COMPATIBLE_VOICE_MODEL=nova
 # TTS_OPEN_AI_COMPATIBLE_ENDPOINT="https://api.openai.com/v1"
 
diff --git a/frontend/src/components/TextToSpeech/OpenAiGenericOptions/index.jsx b/frontend/src/components/TextToSpeech/OpenAiGenericOptions/index.jsx
index 9517c14a9..e579cc292 100644
--- a/frontend/src/components/TextToSpeech/OpenAiGenericOptions/index.jsx
+++ b/frontend/src/components/TextToSpeech/OpenAiGenericOptions/index.jsx
@@ -5,7 +5,7 @@ export default function OpenAiGenericTextToSpeechOptions({ settings }) {
     <div className="w-full flex flex-col gap-y-7">
       <div className="flex gap-x-4">
         <div className="flex flex-col w-60">
-          <div className="flex justify-between items-center mb-2">
+          <div className="flex justify-between items-start mb-2">
             <label className="text-white text-sm font-semibold">Base URL</label>
           </div>
           <input
@@ -23,9 +23,8 @@ export default function OpenAiGenericTextToSpeechOptions({ settings }) {
             will generate TTS responses from.
           </p>
         </div>
-
         <div className="flex flex-col w-60">
-          <label className="text-white text-sm font-semibold block mb-3">
+          <label className="text-white text-sm font-semibold block mb-2">
             API Key
           </label>
           <input
@@ -44,6 +43,28 @@ export default function OpenAiGenericTextToSpeechOptions({ settings }) {
             this is optional if your service does not require one.
           </p>
         </div>
+      </div>
+      <div className="flex gap-x-4">
+        <div className="flex flex-col w-60">
+          <label className="text-white text-sm font-semibold block mb-3">
+            TTS Model
+          </label>
+          <input
+            type="text"
+            name="TTSOpenAICompatibleModel"
+            className="border-none bg-theme-settings-input-bg text-white placeholder:text-theme-settings-input-placeholder text-sm rounded-lg focus:outline-primary-button active:outline-primary-button outline-none block w-full p-2.5"
+            placeholder="Your TTS model identifier"
+            defaultValue={settings?.TTSOpenAICompatibleModel}
+            required={true}
+            autoComplete="off"
+            spellCheck={false}
+          />
+          <p className="text-xs leading-[18px] font-base text-white text-opacity-60 mt-2">
+            Most TTS services will have several models available. This is the{" "}
+            <code>model</code> parameter you will use to select the model you
+            want to use. Note: This is not the same as the voice model.
+          </p>
+        </div>
         <div className="flex flex-col w-60">
           <label className="text-white text-sm font-semibold block mb-3">
             Voice Model
diff --git a/server/.env.example b/server/.env.example
index 809e36ed8..12fa5ec22 100644
--- a/server/.env.example
+++ b/server/.env.example
@@ -257,6 +257,7 @@ TTS_PROVIDER="native"
 
 # TTS_PROVIDER="generic-openai"
 # TTS_OPEN_AI_COMPATIBLE_KEY=sk-example
+# TTS_OPEN_AI_COMPATIBLE_MODEL=tts-1
 # TTS_OPEN_AI_COMPATIBLE_VOICE_MODEL=nova
 # TTS_OPEN_AI_COMPATIBLE_ENDPOINT="https://api.openai.com/v1"
 
diff --git a/server/models/systemSettings.js b/server/models/systemSettings.js
index bb7311fb8..04596e5b7 100644
--- a/server/models/systemSettings.js
+++ b/server/models/systemSettings.js
@@ -256,6 +256,7 @@ const SystemSettings = {
         process.env.TTS_PIPER_VOICE_MODEL ?? "en_US-hfc_female-medium",
       // OpenAI Generic TTS
       TTSOpenAICompatibleKey: !!process.env.TTS_OPEN_AI_COMPATIBLE_KEY,
+      TTSOpenAICompatibleModel: process.env.TTS_OPEN_AI_COMPATIBLE_MODEL,
       TTSOpenAICompatibleVoiceModel:
         process.env.TTS_OPEN_AI_COMPATIBLE_VOICE_MODEL,
       TTSOpenAICompatibleEndpoint: process.env.TTS_OPEN_AI_COMPATIBLE_ENDPOINT,
diff --git a/server/utils/TextToSpeech/openAiGeneric/index.js b/server/utils/TextToSpeech/openAiGeneric/index.js
index df39e6348..5694ed33a 100644
--- a/server/utils/TextToSpeech/openAiGeneric/index.js
+++ b/server/utils/TextToSpeech/openAiGeneric/index.js
@@ -4,6 +4,10 @@ class GenericOpenAiTTS {
       this.#log(
         "No OpenAI compatible API key was set. You might need to set this to use your OpenAI compatible TTS service."
       );
+    if (!process.env.TTS_OPEN_AI_COMPATIBLE_MODEL)
+      this.#log(
+        "No OpenAI compatible TTS model was set. We will use the default voice model 'tts-1'. This may not exist or be valid your selected endpoint."
+      );
     if (!process.env.TTS_OPEN_AI_COMPATIBLE_VOICE_MODEL)
       this.#log(
         "No OpenAI compatible voice model was set. We will use the default voice model 'alloy'. This may not exist for your selected endpoint."
@@ -18,7 +22,11 @@ class GenericOpenAiTTS {
       apiKey: process.env.TTS_OPEN_AI_COMPATIBLE_KEY || null,
       baseURL: process.env.TTS_OPEN_AI_COMPATIBLE_ENDPOINT,
     });
+    this.model = process.env.TTS_OPEN_AI_COMPATIBLE_MODEL ?? "tts-1";
     this.voice = process.env.TTS_OPEN_AI_COMPATIBLE_VOICE_MODEL ?? "alloy";
+    this.#log(
+      `Service (${process.env.TTS_OPEN_AI_COMPATIBLE_ENDPOINT}) with model: ${this.model} and voice: ${this.voice}`
+    );
   }
 
   #log(text, ...args) {
@@ -33,7 +41,7 @@ class GenericOpenAiTTS {
   async ttsBuffer(textInput) {
     try {
       const result = await this.openai.audio.speech.create({
-        model: "tts-1",
+        model: this.model,
         voice: this.voice,
         input: textInput,
       });
diff --git a/server/utils/helpers/updateENV.js b/server/utils/helpers/updateENV.js
index f64a042ce..87670830d 100644
--- a/server/utils/helpers/updateENV.js
+++ b/server/utils/helpers/updateENV.js
@@ -601,6 +601,10 @@ const KEY_MAPPING = {
     envKey: "TTS_OPEN_AI_COMPATIBLE_KEY",
     checks: [],
   },
+  TTSOpenAICompatibleModel: {
+    envKey: "TTS_OPEN_AI_COMPATIBLE_MODEL",
+    checks: [],
+  },
   TTSOpenAICompatibleVoiceModel: {
     envKey: "TTS_OPEN_AI_COMPATIBLE_VOICE_MODEL",
     checks: [isNotEmpty],