Skip to content

Commit c44a81b

Browse files
FChin39FChin39
committed
Update api.py
1 parent 39fc6f2 commit c44a81b

File tree

1 file changed

+79
-42
lines changed

1 file changed

+79
-42
lines changed

‎api.py

+79-42
Original file line numberDiff line numberDiff line change
@@ -14,21 +14,13 @@
1414
from pydantic import BaseModel
1515
import threading
1616
import uvicorn
17+
import logging
1718

18-
from tools.torchgate import TorchGate
19-
import tools.rvc_for_realtime as rvc_for_realtime
20-
from configs.config import Config
21-
22-
load_dotenv()
23-
os.environ["OMP_NUM_THREADS"] = "4"
24-
if sys.platform == "darwin":
25-
os.environ["PYTORCH_ENABLE_MPS_FALLBACK"] = "1"
26-
27-
now_dir = os.getcwd()
28-
sys.path.append(now_dir)
29-
30-
stream_latency = -1
19+
# Initialize the logger
20+
logging.basicConfig(level=logging.INFO)
21+
logger = logging.getLogger(__name__)
3122

23+
# Define FastAPI app
3224
app = FastAPI()
3325

3426
class GUIConfig:
@@ -69,23 +61,25 @@ class ConfigData(BaseModel):
6961
class AudioAPI:
7062
def __init__(self) -> None:
7163
self.gui_config = GUIConfig()
72-
self.config = Config()
64+
self.config = None # Initialize Config object as None
7365
self.flag_vc = False
7466
self.function = "vc"
7567
self.delay_time = 0
68+
self.rvc = None # Initialize RVC object as None
7669

7770
def load(self):
7871
input_devices, output_devices, _, _ = self.get_devices()
7972
try:
80-
with open("configs/config.json", "r") as j:
73+
with open("configs/config.json", "r", encoding='utf-8') as j:
8174
data = json.load(j)
8275
data["rmvpe"] = True # Ensure rmvpe is the only f0method
8376
if data["sg_input_device"] not in input_devices:
8477
data["sg_input_device"] = input_devices[sd.default.device[0]]
8578
if data["sg_output_device"] not in output_devices:
8679
data["sg_output_device"] = output_devices[sd.default.device[1]]
87-
except:
88-
with open("configs/config.json", "w") as j:
80+
except Exception as e:
81+
logger.error(f"Failed to load configuration: {e}")
82+
with open("configs/config.json", "w", encoding='utf-8') as j:
8983
data = {
9084
"pth_path": " ",
9185
"index_path": " ",
@@ -102,18 +96,15 @@ def load(self):
10296
"use_jit": False,
10397
}
10498
data["rmvpe"] = True # Ensure rmvpe is the only f0method
99+
json.dump(data, j, ensure_ascii=False)
105100
return data
106101

107102
def set_values(self, values):
108-
if len(values.pth_path.strip()) == 0:
103+
logger.info(f"Setting values: {values}")
104+
if not values.pth_path.strip():
109105
raise HTTPException(status_code=400, detail="Please select a .pth file")
110-
if len(values.index_path.strip()) == 0:
106+
if not values.index_path.strip():
111107
raise HTTPException(status_code=400, detail="Please select an index file")
112-
pattern = re.compile("[^\x00-\x7F]+")
113-
if pattern.findall(values.pth_path):
114-
raise HTTPException(status_code=400, detail=".pth file path should not contain non-ASCII characters")
115-
if pattern.findall(values.index_path):
116-
raise HTTPException(status_code=400, detail="Index file path should not contain non-ASCII characters")
117108
self.set_devices(values.sg_input_device, values.sg_output_device)
118109
self.config.use_jit = False
119110
self.gui_config.pth_path = values.pth_path
@@ -143,10 +134,8 @@ def start_vc(self):
143134
0,
144135
0,
145136
self.config,
146-
self.rvc if hasattr(self, "rvc") else None,
137+
self.rvc if self.rvc else None,
147138
)
148-
if not hasattr(self.rvc, 'tgt_sr'):
149-
self.rvc.tgt_sr = 44100
150139
self.gui_config.samplerate = self.rvc.tgt_sr
151140
self.zc = self.rvc.tgt_sr // 100
152141
self.block_frame = (
@@ -226,8 +215,8 @@ def soundinput(self):
226215
stream_latency = stream.latency[-1]
227216
while self.flag_vc:
228217
time.sleep(self.gui_config.block_time)
229-
print("Audio block passed.")
230-
print("ENDing VC")
218+
logger.info("Audio block passed.")
219+
logger.info("Ending VC")
231220

232221
def audio_callback(self, indata: np.ndarray, outdata: np.ndarray, frames, times, status):
233222
start_time = time.perf_counter()
@@ -289,7 +278,7 @@ def audio_callback(self, indata: np.ndarray, outdata: np.ndarray, frames, times,
289278
sola_offset = sola_offset.item()
290279
else:
291280
sola_offset = torch.argmax(cor_nom[0, 0] / cor_den[0, 0])
292-
print("sola_offset = %d" % int(sola_offset))
281+
logger.info(f"sola_offset = {sola_offset}")
293282
infer_wav = infer_wav[sola_offset : sola_offset + self.block_frame + self.crossfade_frame]
294283
infer_wav[: self.crossfade_frame] *= self.fade_in_window
295284
infer_wav[: self.crossfade_frame] += self.sola_buffer * self.fade_out_window
@@ -299,7 +288,7 @@ def audio_callback(self, indata: np.ndarray, outdata: np.ndarray, frames, times,
299288
else:
300289
outdata[:] = infer_wav[: -self.crossfade_frame].repeat(2, 1).t().cpu().numpy()
301290
total_time = time.perf_counter() - start_time
302-
print("Infer time: %.2f" % total_time)
291+
logger.info(f"Infer time: {total_time:.2f}")
303292

304293
def get_devices(self, update: bool = True):
305294
if update:
@@ -344,34 +333,61 @@ def set_devices(self, input_device, output_device):
344333
input_device_indices,
345334
output_device_indices,
346335
) = self.get_devices()
336+
logger.debug(f"Available input devices: {input_devices}")
337+
logger.debug(f"Available output devices: {output_devices}")
338+
logger.debug(f"Selected input device: {input_device}")
339+
logger.debug(f"Selected output device: {output_device}")
340+
341+
if input_device not in input_devices:
342+
logger.error(f"Input device '{input_device}' is not in the list of available devices")
343+
raise HTTPException(status_code=400, detail=f"Input device '{input_device}' is not available")
344+
345+
if output_device not in output_devices:
346+
logger.error(f"Output device '{output_device}' is not in the list of available devices")
347+
raise HTTPException(status_code=400, detail=f"Output device '{output_device}' is not available")
348+
347349
sd.default.device[0] = input_device_indices[input_devices.index(input_device)]
348350
sd.default.device[1] = output_device_indices[output_devices.index(output_device)]
349-
print("Input device: %s:%s" % (str(sd.default.device[0]), input_device))
350-
print("Output device: %s:%s" % (str(sd.default.device[1]), output_device))
351+
logger.info(f"Input device set to {sd.default.device[0]}: {input_device}")
352+
logger.info(f"Output device set to {sd.default.device[1]}: {output_device}")
351353

352354
audio_api = AudioAPI()
353355

354-
@app.get("/inputDevices")
356+
@app.get("/inputDevices", response_model=list)
355357
def get_input_devices():
356-
input_devices, _, _, _ = audio_api.get_devices()
357-
return input_devices
358+
try:
359+
input_devices, _, _, _ = audio_api.get_devices()
360+
return input_devices
361+
except Exception as e:
362+
logger.error(f"Failed to get input devices: {e}")
363+
raise HTTPException(status_code=500, detail="Failed to get input devices")
358364

359-
@app.get("/outputDevices")
365+
@app.get("/outputDevices", response_model=list)
360366
def get_output_devices():
361-
_, output_devices, _, _ = audio_api.get_devices()
362-
return output_devices
367+
try:
368+
_, output_devices, _, _ = audio_api.get_devices()
369+
return output_devices
370+
except Exception as e:
371+
logger.error(f"Failed to get output devices: {e}")
372+
raise HTTPException(status_code=500, detail="Failed to get output devices")
363373

364374
@app.post("/config")
365375
def configure_audio(config_data: ConfigData):
366376
try:
377+
logger.info(f"Configuring audio with data: {config_data}")
367378
if audio_api.set_values(config_data):
368379
settings = config_data.dict()
369380
settings["use_jit"] = False
370381
settings["f0method"] = "rmvpe"
371-
with open("configs/config.json", "w") as j:
372-
json.dump(settings, j)
382+
with open("configs/config.json", "w", encoding='utf-8') as j:
383+
json.dump(settings, j, ensure_ascii=False)
384+
logger.info("Configuration set successfully")
373385
return {"message": "Configuration set successfully"}
386+
except HTTPException as e:
387+
logger.error(f"Configuration error: {e.detail}")
388+
raise
374389
except Exception as e:
390+
logger.error(f"Configuration failed: {e}")
375391
raise HTTPException(status_code=400, detail=f"Configuration failed: {e}")
376392

377393
@app.post("/start")
@@ -381,8 +397,13 @@ def start_conversion():
381397
audio_api.start_vc()
382398
return {"message": "Audio conversion started"}
383399
else:
400+
logger.warning("Audio conversion already running")
384401
raise HTTPException(status_code=400, detail="Audio conversion already running")
402+
except HTTPException as e:
403+
logger.error(f"Start conversion error: {e.detail}")
404+
raise
385405
except Exception as e:
406+
logger.error(f"Failed to start conversion: {e}")
386407
raise HTTPException(status_code=500, detail=f"Failed to start conversion: {e}")
387408

388409
@app.post("/stop")
@@ -394,9 +415,25 @@ def stop_conversion():
394415
stream_latency = -1
395416
return {"message": "Audio conversion stopped"}
396417
else:
418+
logger.warning("Audio conversion not running")
397419
raise HTTPException(status_code=400, detail="Audio conversion not running")
420+
except HTTPException as e:
421+
logger.error(f"Stop conversion error: {e.detail}")
422+
raise
398423
except Exception as e:
424+
logger.error(f"Failed to stop conversion: {e}")
399425
raise HTTPException(status_code=500, detail=f"Failed to stop conversion: {e}")
400426

401427
if __name__ == "__main__":
402-
uvicorn.run(app, host="0.0.0.0", port=8043)
428+
if sys.platform == "win32":
429+
from multiprocessing import freeze_support
430+
freeze_support()
431+
load_dotenv()
432+
os.environ["OMP_NUM_THREADS"] = "4"
433+
if sys.platform == "darwin":
434+
os.environ["PYTORCH_ENABLE_MPS_FALLBACK"] = "1"
435+
from tools.torchgate import TorchGate
436+
import tools.rvc_for_realtime as rvc_for_realtime
437+
from configs.config import Config
438+
audio_api.config = Config()
439+
uvicorn.run(app, host="0.0.0.0", port=6242)

0 commit comments

Comments
 (0)