Gokulvarthan-Ramesh/aionboarding
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: implement VoiceService with multi-model speech-to-text and text-to-speech capabilities

Browse Source
This commit is contained in:
Gokulvarthan-Ramesh
2026-08-28 12:57:22 +05:30
parent 1645d9b1a8
commit f66ade7051
1 changed files with 4 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ai-patient-onboarding/app/services/voice_service.py
+4 -4
View File
@@ -121,8 +121,8 @@ class VoiceService:
and lang_prefix in ["ta", "hi", "te", "kn", "bn", "mr", "gu", "ml", "pa", "od", "en"] and lang_prefix in ["ta", "hi", "te", "kn", "bn", "mr", "gu", "ml", "pa", "od", "en"]
): ):
try: try:
speaker = "anushka" if lang_prefix in ["ta", "hi", "te", "kn"] else "karun" speaker = "kavya" if lang_prefix in ["ta", "hi", "te", "kn"] else "gokul"
async with httpx2.AsyncClient(timeout=3.0) as client: async with httpx2.AsyncClient(timeout=5.0) as client:
headers = { headers = {
"api-subscription-key": self.sarvam_key, "api-subscription-key": self.sarvam_key,
"Content-Type": "application/json", "Content-Type": "application/json",
@@ -131,7 +131,7 @@ class VoiceService:
"inputs": [cleaned_text], "inputs": [cleaned_text],
"target_language_code": language_code, "target_language_code": language_code,
"speaker": speaker, "speaker": speaker,
"model": "bulbul:v2", "model": "bulbul:v3",
} }
res = await client.post( res = await client.post(
"https://api.sarvam.ai/text-to-speech", "https://api.sarvam.ai/text-to-speech",
@@ -153,7 +153,7 @@ class VoiceService:
# 2. Use OpenAI TTS # 2. Use OpenAI TTS
if self.openai_key and self.openai_key.strip() not in ["your_openai_api_key", "your_openai_key", ""]: if self.openai_key and self.openai_key.strip() not in ["your_openai_api_key", "your_openai_key", ""]:
client = AsyncOpenAI(api_key=self.openai_key, timeout=1.5) client = AsyncOpenAI(api_key=self.openai_key, timeout=5.0)
try: try:
response = await client.audio.speech.create( response = await client.audio.speech.create(
model="tts-1", model="tts-1",