Spaces:
Building
Building
File size: 8,417 Bytes
bea4a17 5d32118 bea4a17 5d32118 bea4a17 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 |
"""Audio API endpoints for Flare
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Provides text-to-speech (TTS) and speech-to-text (STT) endpoints.
"""
from fastapi import APIRouter, HTTPException, Response, Body
from pydantic import BaseModel
from typing import Optional
from datetime import datetime
import sys
from logger import log_info, log_error, log_warning, log_debug
from tts_factory import TTSFactory
from tts_preprocessor import TTSPreprocessor
from config_provider import ConfigProvider
router = APIRouter(tags=["audio"])
# ===================== Models =====================
class TTSRequest(BaseModel):
text: str
voice_id: Optional[str] = None
language: Optional[str] = "tr-TR"
class STTRequest(BaseModel):
audio_data: str # Base64 encoded audio
language: Optional[str] = "tr-TR"
format: Optional[str] = "webm" # webm, wav, mp3
# ===================== Helpers =====================
def log(message: str):
"""Log helper with timestamp"""
timestamp = datetime.now().strftime("%H:%M:%S.%f")[:-3]
print(f"[{timestamp}] {message}")
sys.stdout.flush()
# ===================== TTS Endpoints =====================
@router.post("/tts/generate")
async def generate_tts(request: TTSRequest):
"""Generate TTS audio from text - public endpoint for chat"""
try:
# Create TTS provider
tts_provider = TTSFactory.create_provider()
if not tts_provider:
# Return empty response for no TTS
log_info("📵 TTS disabled - returning empty response")
return Response(
content=b"",
media_type="audio/mpeg",
headers={"X-TTS-Status": "disabled"}
)
log_info(f"🎤 TTS request: '{request.text[:50]}...' with provider: {tts_provider.get_provider_name()}")
# Preprocess text if needed
preprocessor = TTSPreprocessor(language=request.language)
processed_text = preprocessor.preprocess(
request.text,
tts_provider.get_preprocessing_flags()
)
log_debug(f"📝 Preprocessed text: {processed_text[:100]}...")
# Generate audio
audio_data = await tts_provider.synthesize(
text=processed_text,
voice_id=request.voice_id
)
log_info(f"✅ TTS generated {len(audio_data)} bytes of audio")
# Return audio as binary response
return Response(
content=audio_data,
media_type="audio/mpeg",
headers={
"Content-Disposition": 'inline; filename="tts_output.mp3"',
"X-TTS-Provider": tts_provider.get_provider_name(),
"X-TTS-Language": request.language,
"Cache-Control": "no-cache"
}
)
except Exception as e:
log_error("❌ TTS generation error", e)
raise HTTPException(
status_code=500,
detail=f"TTS generation failed: {str(e)}"
)
@router.get("/tts/voices")
async def get_tts_voices():
"""Get available TTS voices - public endpoint"""
try:
tts_provider = TTSFactory.create_provider()
if not tts_provider:
return {
"voices": [],
"provider": "none",
"enabled": False
}
voices = tts_provider.get_supported_voices()
# Convert dict to list format
voice_list = [
{"id": voice_id, "name": voice_name}
for voice_id, voice_name in voices.items()
]
return {
"voices": voice_list,
"provider": tts_provider.get_provider_name(),
"enabled": True
}
except Exception as e:
log_error("❌ Error getting TTS voices", e)
return {
"voices": [],
"provider": "error",
"enabled": False,
"error": str(e)
}
@router.get("/tts/status")
async def get_tts_status():
"""Get TTS service status"""
cfg = ConfigProvider.get()
return {
"enabled": cfg.global_config.tts_provider.name != "no_tts",
"provider": cfg.global_config.tts_provider.name,
"provider_config": {
"name": cfg.global_config.tts_provider.name,
"has_api_key": bool(cfg.global_config.tts_provider.api_key),
"endpoint": cfg.global_config.tts_provider.endpoint
}
}
# ===================== STT Endpoints =====================
@router.post("/stt/transcribe")
async def transcribe_audio(request: STTRequest):
"""Transcribe audio to text"""
try:
from stt_factory import STTFactory
from stt_interface import STTConfig
import base64
# Create STT provider
stt_provider = STTFactory.create_provider()
if not stt_provider or not stt_provider.supports_realtime():
log_warning("📵 STT disabled or doesn't support transcription")
raise HTTPException(
status_code=503,
detail="STT service not available"
)
# Get config
cfg = ConfigProvider.get()
stt_config = cfg.global_config.stt_provider.settings
# Decode audio data
audio_bytes = base64.b64decode(request.audio_data)
# Create STT config
config = STTConfig(
language=request.language or stt_config.get("language", "tr-TR"),
sample_rate=16000,
encoding=request.format.upper() if request.format else "WEBM_OPUS",
enable_punctuation=stt_config.get("enable_punctuation", True),
enable_word_timestamps=False,
model=stt_config.get("model", "latest_long"),
use_enhanced=stt_config.get("use_enhanced", True),
single_utterance=True,
interim_results=False
)
# Start streaming session
await stt_provider.start_streaming(config)
# Process audio
transcription = ""
confidence = 0.0
try:
async for result in stt_provider.stream_audio(audio_bytes):
if result.is_final:
transcription = result.text
confidence = result.confidence
break
finally:
# Stop streaming
await stt_provider.stop_streaming()
log_info(f"✅ STT transcription completed: '{transcription[:50]}...'")
return {
"text": transcription,
"confidence": confidence,
"language": request.language,
"provider": stt_provider.get_provider_name()
}
except HTTPException:
raise
except Exception as e:
log_error("❌ STT transcription error", e)
raise HTTPException(
status_code=500,
detail=f"Transcription failed: {str(e)}"
)
@router.get("/stt/languages")
async def get_stt_languages():
"""Get supported STT languages"""
try:
from stt_factory import STTFactory
stt_provider = STTFactory.create_provider()
if not stt_provider:
return {
"languages": [],
"provider": "none",
"enabled": False
}
languages = stt_provider.get_supported_languages()
return {
"languages": languages,
"provider": stt_provider.get_provider_name(),
"enabled": True
}
except Exception as e:
log_error("❌ Error getting STT languages", e)
return {
"languages": [],
"provider": "error",
"enabled": False,
"error": str(e)
}
@router.get("/stt/status")
async def get_stt_status():
"""Get STT service status"""
cfg = ConfigProvider.get()
return {
"enabled": cfg.global_config.stt_provider.name != "no_stt",
"provider": cfg.global_config.stt_provider.name,
"provider_config": {
"name": cfg.global_config.stt_provider.name,
"has_api_key": bool(cfg.global_config.stt_provider.api_key),
"endpoint": cfg.global_config.stt_provider.endpoint
}
} |