5334ea0f7a
Add WebSocket authentication via api_key
...
- When --api_key is set, WebSocket connections require auth too
- Supports Authorization: Bearer <key> header or ?token=<key> query param
- Unauthenticated connections receive HTTP 401 before upgrade
- Uses websockets process_request callback (no resource allocation before auth)
- Added 5 unit tests for WebSocket auth handler
2026-05-26 23:07:50 -04:00
b648bcb2a4
Add optional API key auth and rate limiting for REST API
...
- api_key param: requires 'Authorization: Bearer <key>' header
- rate_limit_rpm param: per-IP sliding-window rate limit (requests/min)
- Both are off by default (backward compatible)
- CLI flags: --api_key, --rate_limit_rpm
- Added 5 unit tests for auth and rate limiting
2026-05-26 23:02:43 -04:00
ced4bdb737
feat: add Prometheus metrics instrumentation
...
- New whisper_live/metrics.py with Counter, Gauge, Histogram metrics
- Track connections (opened/closed/rejected), transcription latency,
audio processed, segments emitted, REST requests, and errors
- All metric helpers are no-ops when prometheus_client not installed
- --metrics_port CLI flag to expose /metrics endpoint (0 = disabled)
- Metrics integrated into server.py, base.py at key instrumentation points
- 17 new tests in tests/test_metrics.py (178 total passing)
2026-05-13 10:45:40 -04:00
f5340ddf1e
audio: add support for raw pcm input via server flag
...
fixes #
2026-04-17 09:21:06 -04:00
e8bd4fd532
Add cross-client GPU batch inference for faster_whisper backend
2026-02-26 00:00:51 +00:00
29ee640409
api: add support OpenAI REST transcription api
2026-02-05 22:31:59 -05:00
95a9b7ef05
fix(run_server.py): help text for max_connection_time argument
...
The help text for `--max_connection_time` is incorrect. Looks like a copy-paste mistake from `--cache_path`.
2025-09-25 11:14:35 -03:00
8d6ddd4f7b
Change max_clients max_connection_time from server only
...
Signed-off-by: makaveli10 <vineet.suryan@collabora.com >
2025-07-21 22:52:24 +05:30
74abf66d48
Make cache path configurable to save auto converted ct2 models
...
Signed-off-by: makaveli10 <vineet.suryan@collabora.com >
2025-06-01 09:04:22 +00:00
47ee035f65
Upgrade tensorrt_llm to v0.18.2
...
Signed-off-by: makaveli10 <vineet.suryan@collabora.com >
2025-04-22 12:33:03 +00:00
a62495b090
Integrate OpenVINO backend
...
Signed-off-by: makaveli <vineet.suryan@collabora.com >
2025-03-31 12:57:19 +05:30
14077315ae
Fix argparser option
2024-06-05 10:34:22 +02:00
ab17c4dbc6
Make single model mode the default, update readme
2024-06-05 09:47:52 +02:00
3c09289dea
Add single model mode for custom models
...
- Use a threadlock around the model in single model mode
2024-05-31 15:06:47 +02:00
819ab35b28
fix: limit CPU usage for VAD onnxruntime inference session by setting OMP_NUM_THREADS
...
Signed-off-by: makaveli10 <vineet.suryan@collabora.com >
2024-05-24 04:58:49 -04:00
9fbff47126
🔨 refactor whisper_live according to flake8
2024-02-09 13:45:13 +05:30
735d6c7763
merge with main
2024-01-19 11:43:21 +00:00
647c576e6a
update with multilingual option
2024-01-11 08:17:56 +00:00
244ca9e6ba
remove duplicate code
2024-01-10 14:14:48 +00:00
0f9e93d203
add: tensorrt backend to server
2024-01-09 18:10:17 +00:00
9e60160f87
remove unused improt
2023-08-09 04:06:11 +08:00
1a4775bac2
add client queue
2023-08-07 22:07:51 +08:00
8db533dd6b
Use the updated run method to start the server.
2023-08-03 09:54:14 -04:00
f6f5b6ba8a
run_server to start server in docker
2023-08-02 16:16:48 +08:00
Aaron Boxer