CoolFace
Modelpublic

cstr/firered-lid-GGUF

sourceHugging Faceapache-2.0updated 2mo agoView on Hugging Face
1likes764downloads
Model Card

FireRedLID -- GGUF

GGUF conversion of `FireRedTeam/FireRedLID` for use with [CrispStrobe/CrispASR](https://github.com/CrispStrobe/CrispASR).

Available variants

FileTypeSizeNotes
firered-lid.ggufF161.7 GBEncoder (F16) + LID decoder

Model details

  • —Architecture: Conformer encoder (16L, d=1280, 20 heads) + 6L Transformer decoder (8 heads, cross-attention)
  • —Parameters: ~887M (723M shared encoder + 164M LID decoder)
  • —Languages: 100+ languages + 20+ Chinese dialects
  • —Accuracy: 97.18% on FLEURS-82, 92.07% on CommonVoice-74
  • —License: Apache 2.0
  • —Output: Language code (e.g., "en", "zh", "mandarin", "ja")

Usage with CrispASR

bash
# Language identification
./build/bin/crispasr --backend firered-asr -m firered-lid.gguf -f audio.wav

The encoder is shared with FireRedASR2-AED. The LID decoder uses 8 attention heads (vs 20 for ASR).

Conversion

bash
python models/convert-firered-lid-to-gguf.py --input FireRedTeam/FireRedLID --output firered-lid.gguf

Provenance and EU AI Act Art. 53 note

  • —Upstream model: FireRedTeam/FireRedLID — published by FireRedTeam.
  • —Upstream licence: apache-2.0. This repository redistributes under the same terms; it grants no rights the upstream licence does not.
  • —What was done here: format conversion and/or quantisation only (GGUF/GGML). No training, no fine-tuning, no merging, no distillation, no change to architecture, vocabulary or capability. Only the numeric representation of the upstream weights differs.
  • —Training data: documented — where it is documented at all — by the upstream provider; see the upstream model card. No training data was used, added or selected by this repository. No training-content summary was found on the upstream model card at the time of writing; that documentation gap is upstream's and is not filled here.
  • —Provider status: under Regulation (EU) 2024/1689 the upstream authors remain the provider of this model. Converting the serialisation format does not make this repository the provider of a new general-purpose AI model, and no such claim is made. Questions about training content, copyright policy or model capability belong upstream.