Skip to main content
Version: 2026.03.0-rc1

Phonexia Language identification changelog

2.0.1 (2026-01-13)​

Fixed​

  • [PLAT-1506] Fixed a rare issue where audio resampling could fail when processing large files, particularly when converting between certain sample rates (e.g., 11025 Hz to 8000 Hz).
  • [PLAT-1509] Fixed GPU processing failure on NVIDIA Blackwell architecture GPUs (e.g., RTX 1000 series) caused by a bug in ONNX Runtime.
  • Fixed license validation error handling for legacy (v1) license files with missing required fields. Previously, this could cause undefined behavior.

2.0.0 (2025-11-24)​

Breaking change​

  • [PLAT-1425] Improve compatibility with future models - WARNING microservice requires at least model v5.5.0

Changed​

  • Microservice now by default logs each request with incremental correlation id
  • [PLAT-1471] Refactoring of the internal audio handling to reduce memory usage. Audio format support is now limited to WAV and FLAC.

Fixed​

  • [VOX-2666] Inconsistent logging
  • [PLAT-1277] Possible memory leak on cuda device

1.7.0 (2025-04-28)​

Added​

  • [PLAT-1056] Support for new licenses tied only to model major version. Old licenses tied to precise model version are still supported.
  • [PLAT-1070] Default port 8080 exposed in Dockerfile
  • [VOX-1293] phonexia.grpc.common.Status service
  • [VOX-1293] Field license_flags to LicensingInfoResult in phonexia.grpc.common.Licensing

Fixed​

  • [VOX-1497] Crash when recording sample rate is not divisible by 10 and processing start time has been set

1.6.1 (2024-12-10)​

Fixed​

  • [PLAT-978] Loading configuration from model

1.6.0 (2024-11-21)​

Added​

  • [PLAT-952] Added PHX_NUM_THREADS_PER_INSTANCE, PHX_NUM_INSTANCES_PER_DEVICE and PHX_DEVICE_INDEX options for concurrent processing

Changed​

  • [PLAT-389] Start and end request's messages are logged at the INFO level

1.5.0 (2024-10-31)​

Fixed​

  • [PLAT-827] Delay to response when identify configuration is invalid (configuration now checked at the beginning)
  • [PLAT-942] NaN scores on empty recordings

Added​

  • [PLAT-926] Extract method for languageprint extraction
  • [PLAT-926] Adapt method for adding or modifying languages using adaptation profile

Changed​

  • [PLAT-926] Identify method now accepts languageprints and adaptation profiles
  • [PLAT-926] ListSupportedLanguages method now accepts adaptation profiles

1.4.1 (2024-10-01)​

Fixed​

  • [PLAT-936] Crash when using negative values in time range

1.4.0 (2024-09-12)​

Added​

  • [PLAT-888] Added PHX_KEEPALIVE_TIME_S and PHX_KEEPALIVE_TIMEOUT_S options for connection timeout specification

1.3.1 (2024-09-04)​

Fixed​

  • [PLAT-886] Unexpected application crash on some empty audio with specified groups

1.3.0 (2024-08-20)​

Added​

  • [PLAT-854] Possibility to specify prior weights of languages from a model configuration file

Changed​

  • [PLAT-832] Update chunking algorithm for better accuracy

Fixed​

  • [PLAT-813] Language tags are case insensitive now

1.2.0 (2024-07-10)​

Added​

  • [PLAT-771] Support for RAW audio and real-time audio processing during transmission using the RAW audio
  • [PLAT-781] Support for license extensions
  • [PLAT-812] Add identifier type (whether result is language or group) to the response

Fixed​

  • [PLAT-758] Device is not logged by default

1.1.1 (2024-06-25)​

Fixed​

  • [PLAT-786] Memory consumption spikes for short audio on GPU image

1.1.0 (2024-06-19)​

Changed​

  • [PLAT-791] Groups in the output now encapsulates all the languages it is comprised of

Added​

  • [PLAT-800] Added temperature for final scores scaling

1.0.1 (2024-06-06)​

Fixed​

  • [PLAT-788] Missing entrypoint in GPU docker image

1.0.0 (2024-05-30)​

Added​

  • [PLAT-497] New generation of Phonexia Language Identification