From 5eabd2493c8cd16cc368be569d5781e3234c0069 Mon Sep 17 00:00:00 2001
From: Marco Beretta <81851188+berry-13@users.noreply.github.com>
Date: Sat, 23 Nov 2024 12:20:17 +0100
Subject: [PATCH] =?UTF-8?q?=F0=9F=8C=8A=20feat:=20update=20Deepgram=20SDK?=
 =?UTF-8?q?=20integration=20for=20STT=20and=20remove=20unused=20TTS=20prov?=
 =?UTF-8?q?ider?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 api/server/services/Files/Audio/STTService.js | 12 ++++-
 api/server/services/Files/Audio/TTSService.js | 46 -------------------
 2 files changed, 10 insertions(+), 48 deletions(-)

diff --git a/api/server/services/Files/Audio/STTService.js b/api/server/services/Files/Audio/STTService.js
index bb32ef1b93..23c53a14b6 100644
--- a/api/server/services/Files/Audio/STTService.js
+++ b/api/server/services/Files/Audio/STTService.js
@@ -158,7 +158,15 @@ class STTService {
     return [url, formData, { ...headers, ...formData.getHeaders() }];
   }
 
-  async deepgramSDKProvider(sttSchema, audioReadStream, audioFile) {
+  /**
+   * Transcribes audio using the Deepgram SDK.
+   * @async
+   * @param {Object} sttSchema - The STT schema for Deepgram.
+   * @param {Stream} audioReadStream - The audio data to be transcribed.
+   * @returns {Promise<string>} A promise that resolves to the transcribed text.
+   * @throws {Error} If the transcription fails.
+   */
+  async deepgramSDKProvider(sttSchema, audioReadStream) {
     const apiKey = extractEnvVariable(sttSchema.apiKey) || '';
     const deepgram = createClient(apiKey);
 
@@ -194,7 +202,7 @@ class STTService {
     [configOptions].forEach(this.removeUndefined);
 
     const { result, error } = await deepgram.listen.prerecorded.transcribeFile(
-      Buffer.isBuffer(audioFile) ? audioFile : audioReadStream,
+      audioReadStream,
       configOptions,
     );
 
diff --git a/api/server/services/Files/Audio/TTSService.js b/api/server/services/Files/Audio/TTSService.js
index 8558a7eb9c..5590c8ac7d 100644
--- a/api/server/services/Files/Audio/TTSService.js
+++ b/api/server/services/Files/Audio/TTSService.js
@@ -248,52 +248,6 @@ class TTSService {
     return [url, data, headers];
   }
 
-  deepgramProvider(ttsSchema, input, voice) {
-    const baseUrl = ttsSchema?.url || 'https://api.deepgram.com/v1/speak';
-    const params = {
-      model: ttsSchema.model,
-      voice: voice,
-      language: ttsSchema.language,
-    };
-
-    const queryParams = Object.entries(params)
-      .filter(([, value]) => value)
-      .map(([key, value]) => `${key}=${value}`)
-      .join('&');
-
-    const url = queryParams ? `${baseUrl}?${queryParams}` : baseUrl;
-
-    if (
-      ttsSchema?.voices &&
-      ttsSchema.voices.length > 0 &&
-      !ttsSchema.voices.includes(voice) &&
-      !ttsSchema.voices.includes('ALL')
-    ) {
-      throw new Error(`Voice ${voice} is not available.`);
-    }
-
-    const data = {
-      input,
-      model: ttsSchema?.voices && ttsSchema.voices.length > 0 ? voice : undefined,
-      language: ttsSchema?.language,
-      media_settings: {
-        bit_rate: ttsSchema?.media_settings?.bit_rate,
-        sample_rate: ttsSchema?.media_settings?.sample_rate,
-      },
-    };
-
-    const headers = {
-      'Content-Type': 'application/json',
-      Authorization: `Bearer ${extractEnvVariable(ttsSchema?.apiKey)}`,
-    };
-
-    if (extractEnvVariable(ttsSchema.apiKey) === '') {
-      delete headers.Authorization;
-    }
-
-    return [url, data, headers];
-  }
-
   /**
    * Sends a TTS request to the specified provider.
    * @async