Views
No views yet
voice-detect backend of LocalAI..gguf here is a self-contained, metadata-driven model produced by
voice-detect.cpp, a standalone
C++/ggml engine for speaker recognition and voice analysis. No Python, PyTorch, or
ONNX runtime is needed at inference time: the GGUF carries the weights verbatim plus
the forward-graph topology in its KV metadata, and the C++ engine replays it. One
Kaldi-compatible 80-dim FBank front end feeds every speaker encoder; only the encoder
graph changes between models.1db1759572c90faef6f3a78c36b5941a096a9f89 (voice-detect.cpp)general.architecture = "voicedetect"voice-detect backend| Model | Source | License | Commercial use |
|---|---|---|---|
ecapa-tdnn-voxceleb.gguf | SpeechBrain spkrec-ecapa-voxceleb | Apache-2.0 | Yes |
wespeaker-resnet34-voxceleb.gguf | WeSpeaker VoxCeleb ResNet34 | CC-BY-4.0 | Yes (with attribution) |
eres2net-base-zh-cn.gguf | 3D-Speaker ERes2Net base | Apache-2.0 | Yes |
campplus-zh-cn.gguf | 3D-Speaker CAM++ | Apache-2.0 | Yes |
emotion-wav2vec2-superb-er.gguf | audeering wav2vec2 (emotion) | CC-BY-NC-SA-4.0 | No (research only) |
age-gender-wav2vec2-audeering.gguf | audeering wav2vec2-large-robust (age/gender) | CC-BY-NC-SA-4.0 | No (research only) |
The two wav2vec2 analysis heads (emotion, age/gender) are redistributed under their upstream CC-BY-NC-SA-4.0 terms and are for research / non-commercial use only. The four speaker encoders (ECAPA-TDNN, WeSpeaker, ERes2Net, CAM++) are commercial-friendly.
ecapa-tdnn-voxceleb.gguf - ECAPA-TDNN, 192-dspkrec-ecapa-voxceleb, an
ECAPA-TDNN trained on VoxCeleb, producing a 192-d L2-normalized speaker embedding.
License: Apache-2.0.wespeaker-resnet34-voxceleb.gguf - ResNet34, 256-deres2net-base-zh-cn.gguf - ERes2Net, 512-dcampplus-zh-cn.gguf - CAM++, 192-demotion-wav2vec2-superb-er.gguf - wav2vec2 emotion headanalyze path, not an embedding). License: CC-BY-NC-SA-4.0, research only.age-gender-wav2vec2-audeering.gguf - wav2vec2 age/gender headwav2vec2-large-robust-24-ft-age-gender: a wav2vec2 transformer returning
an age estimate and a gender distribution for an utterance. License: CC-BY-NC-SA-4.0,
research only.| Model | Dtype | Embedding cosine | Result |
|---|---|---|---|
ecapa-tdnn-voxceleb.gguf | f32 | 1.000000 | PASS |
wespeaker-resnet34-voxceleb.gguf | f32 | 1.000000 | PASS |
eres2net-base-zh-cn.gguf | f32 | 1.000000 | PASS |
campplus-zh-cn.gguf | f32 | 1.000000 | PASS |
emotion-wav2vec2-superb-er.gguf | f32 | 1.000000 | PASS |
age-gender-wav2vec2-audeering.gguf | f32 | 1.000000 | PASS |
voice-detect-ecapa-tdnn, -wespeaker-resnet34, -eres2net, -campplus,
-emotion-wav2vec2, -age-gender-wav2vec2) and run by the voice-detect backend,
which exposes speaker embed / verify / identify and analyze (age, gender,
emotion). See the LocalAI and
voice-detect.cpp documentation.