89 lines
3.1 KiB
Python
89 lines
3.1 KiB
Python
from __future__ import annotations
|
|
|
|
import argparse
|
|
import base64
|
|
import json
|
|
import math
|
|
import struct
|
|
import sys
|
|
import urllib.error
|
|
import urllib.request
|
|
import wave
|
|
from io import BytesIO
|
|
|
|
|
|
def post_json(url: str, payload: dict, timeout: int) -> dict:
|
|
request = urllib.request.Request(
|
|
url,
|
|
data=json.dumps(payload, ensure_ascii=False).encode("utf-8"),
|
|
headers={"Content-Type": "application/json"},
|
|
method="POST",
|
|
)
|
|
with urllib.request.urlopen(request, timeout=timeout) as response:
|
|
data = json.loads(response.read().decode("utf-8"))
|
|
if not isinstance(data, dict):
|
|
raise ValueError(f"JSON object expected from {url}")
|
|
return data
|
|
|
|
|
|
def make_wav_data_url(seconds: float = 1.0, sample_rate: int = 16000) -> str:
|
|
frames = int(seconds * sample_rate)
|
|
buffer = BytesIO()
|
|
with wave.open(buffer, "wb") as wav:
|
|
wav.setnchannels(1)
|
|
wav.setsampwidth(2)
|
|
wav.setframerate(sample_rate)
|
|
for index in range(frames):
|
|
envelope = 0.5 if 0.1 * sample_rate < index < 0.9 * sample_rate else 0.0
|
|
sample = int(9000 * envelope * math.sin(2 * math.pi * 440 * index / sample_rate))
|
|
wav.writeframesraw(struct.pack("<h", sample))
|
|
return "data:audio/wav;base64," + base64.b64encode(buffer.getvalue()).decode("ascii")
|
|
|
|
|
|
def run(args: argparse.Namespace) -> int:
|
|
payload = {
|
|
"base_url": args.audio_base_url,
|
|
"audio_base64": make_wav_data_url(args.seconds),
|
|
"filename": "smoke-tone.wav",
|
|
"language": args.language,
|
|
"task": args.task,
|
|
}
|
|
result = post_json(f"{args.chat_base_url.rstrip('/')}/api/audio/transcribe", payload, args.timeout)
|
|
print(json.dumps(result, ensure_ascii=False, indent=2))
|
|
if result.get("error"):
|
|
return 1
|
|
return 0 if result.get("model") and result.get("bytes") else 1
|
|
|
|
|
|
def format_url_error(exc: urllib.error.URLError) -> str:
|
|
if isinstance(exc, urllib.error.HTTPError):
|
|
try:
|
|
body = exc.read().decode("utf-8", errors="replace").strip()
|
|
except OSError:
|
|
body = ""
|
|
return f"HTTP {exc.code}: {body or exc.reason}"
|
|
return str(exc)
|
|
|
|
|
|
def main() -> int:
|
|
parser = argparse.ArgumentParser(description="Smoke test audio transcription through the model chat API.")
|
|
parser.add_argument("--chat-base-url", default="http://192.168.220.91:8765")
|
|
parser.add_argument("--audio-base-url", default="http://docker-gpu.cin.su:8020")
|
|
parser.add_argument("--language", default="russian")
|
|
parser.add_argument("--task", default="transcribe")
|
|
parser.add_argument("--seconds", type=float, default=1.0)
|
|
parser.add_argument("--timeout", type=int, default=600)
|
|
parser.add_argument("--allow-unavailable", action="store_true")
|
|
args = parser.parse_args()
|
|
|
|
try:
|
|
return run(args)
|
|
except (OSError, ValueError, urllib.error.URLError) as exc:
|
|
message = format_url_error(exc) if isinstance(exc, urllib.error.URLError) else str(exc)
|
|
print(f"Smoke audio transcription failed: {message}", file=sys.stderr)
|
|
return 0 if args.allow_unavailable else 1
|
|
|
|
|
|
if __name__ == "__main__":
|
|
raise SystemExit(main())
|