Sponsored by APIMart
Toolify Toolify logo
AI Models
geolid_vl107only_shared_trainable

espnet / geolid_vl107only_shared_trainable

Run on huggingface.co
huggingface.co
Total runs: 18
24-hour runs: 0
7-day runs: 2
30-day runs: 5
Model's Last Updated: September 18 2026
audio-classification

Introduction of geolid_vl107only_shared_trainable

Model Details of geolid_vl107only_shared_trainable

ESPnet2 Spoken Language Identification (LID) model
espnet/geolid_vl107only_shared_trainable

This geolocation-aware language identification (LID) model is developed using the ESPnet toolkit. It integrates the powerful pretrained MMS-1B as the encoder and employs ECAPA-TDNN as the embedding extractor to achieve robust spoken language identification.

The main innovations of this model are:

  1. Incorporating geolocation prediction as an auxiliary task during training.
  2. Conditioning the intermediate representations of the self-supervised learning (SSL) encoder on intermediate-layer information. This geolocation-aware strategy greatly improves robustness, especially for dialects and accented variations.

For further details on the geolocation-aware LID methodology, please refer to our paper: Geolocation-Aware Robust Spoken Language Identification (arXiv link to be added).

Usage Guide: How to use in ESPnet2
Prerequisites

First, ensure you have ESPnet installed. If not, follow the ESPnet installation instructions .

Quick Start

Run the following commands to set up and use the pre-trained model: 

cd espnet

pip install -e .
cd egs2/geolid/lid1

# Download the exp_combined to egs2/geolid/lid1
hf download espnet/geolid_vl107only_shared_trainable --local-dir . --exclude "README.md" "meta.yaml" ".gitattributes"

./run_voxlingua107_only.sh --skip_data_prep false --skip_train true --lid_config conf/voxlingua107_only/mms_ecapa_upcon_32_44_it0.4_shared_trainable.yaml

This will download the pre-trained model and run inference using the VoxLingua107 test data.

Train and Evaluation Datasets

The training used only the VoxLingua107 dataset, comprising 6,628 hours of speech across 107 languages from YouTube.

Dataset Domain #Langs. Train/Test Dialect Training Setup (VL107-only)
VoxLingua107 YouTube 107/33 No Seen
Babel Telephone 25/25 No Unseen
FLEURS Read speech 102/102 No Unseen
ML-SUPERB 2.0 Mixed 137/(137, 8) Yes Unseen
VoxPopuli Parliament 16/16 No Unseen
Results

Accuracy (%) on In-domain and Out-of-domain Test Sets

ESPnet Recipe Config VoxLingua107 Babel FLEURS ML-SUPERB2.0 Dev ML-SUPERB2.0 Dialect VoxPopuli Macro Avg.
conf/voxlingua107_only/mms_ecapa_upcon_32_44_it0.4_shared_trainable.yaml
94.9 87.7 93.5 89.3 78.8 89.5 88.9

For more detailed inference results, please refer to the exp_voxlingua107_only/lid_mms_ecapa_upcon_32_44_it0.4_shared_trainable_raw/inference directory in this repository.

Note (2025-08-18):
The corresponding GitHub recipe egs2/geolid/lid1 has not yet been merged into the ESPnet master branch.
See TODO: add PR link for the latest updates.

LID config
expand 
config: conf/voxlingua107_only/mms_ecapa_upcon_32_44_it0.4_shared_trainable.yaml
print_config: false
log_level: INFO
drop_last_iter: false
dry_run: false
iterator_type: category
valid_iterator_type: category
output_dir: exp_voxlingua107_only/lid_mms_ecapa_upcon_32_44_it0.4_shared_trainable_raw
ngpu: 1
seed: 3702
num_workers: 8
num_att_plot: 0
dist_backend: nccl
dist_init_method: env://
dist_world_size: null
dist_rank: null
local_rank: 0
dist_master_addr: null
dist_master_port: null
dist_launcher: null
multiprocessing_distributed: false
unused_parameters: true
sharded_ddp: false
use_deepspeed: false
deepspeed_config: null
gradient_as_bucket_view: true
ddp_comm_hook: null
cudnn_enabled: true
cudnn_benchmark: true
cudnn_deterministic: false
use_tf32: false
collect_stats: false
write_collected_feats: false
max_epoch: 30
patience: null
val_scheduler_criterion:
- valid
- loss
early_stopping_criterion:
- valid
- loss
- min
best_model_criterion:
-   - valid
    - accuracy
    - max
keep_nbest_models: 2
nbest_averaging_interval: 0
grad_clip: 9999
grad_clip_type: 2.0
grad_noise: false
accum_grad: 2
no_forward_run: false
resume: true
train_dtype: float32
use_amp: true
log_interval: 100
use_matplotlib: true
use_tensorboard: true
create_graph_in_tensorboard: false
use_wandb: false
wandb_project: null
wandb_id: null
wandb_entity: null
wandb_name: null
wandb_model_log_interval: -1
detect_anomaly: false
use_adapter: false
adapter: lora
save_strategy: all
adapter_conf: {}
pretrain_path: null
init_param: []
ignore_init_mismatch: false
freeze_param: []
num_iters_per_epoch: 1000
batch_size: 20
valid_batch_size: null
batch_bins: 2880000
valid_batch_bins: null
category_sample_size: 10
upsampling_factor: 0.5
category_upsampling_factor: 0.5
dataset_upsampling_factor: 0.5
dataset_scaling_factor: 1.2
max_batch_size: 16
min_batch_size: 1
train_shape_file:
- exp_voxlingua107_only/lid_stats_16k/train/speech_shape
valid_shape_file:
- exp_voxlingua107_only/lid_stats_16k/valid/speech_shape
batch_type: catpow
language_upsampling_factor: 0.5
valid_batch_type: null
fold_length:
- 120000
sort_in_batch: descending
shuffle_within_batch: false
sort_batch: descending
multiple_iterator: false
chunk_length: 500
chunk_shift_ratio: 0.5
num_cache_chunks: 1024
chunk_excluded_key_prefixes: []
chunk_default_fs: null
chunk_max_abs_length: null
chunk_discard_short_samples: true
train_data_path_and_name_and_type:
-   - dump/raw/train_voxlingua107_lang/wav.scp
    - speech
    - sound
-   - dump/raw/train_voxlingua107_lang/utt2lang
    - lid_labels
    - text
valid_data_path_and_name_and_type:
-   - dump/raw/dev_voxlingua107_lang/wav.scp
    - speech
    - sound
-   - dump/raw/dev_voxlingua107_lang/utt2lang
    - lid_labels
    - text
multi_task_dataset: false
allow_variable_data_keys: false
max_cache_size: 0.0
max_cache_fd: 32
allow_multi_rates: false
valid_max_cache_size: null
exclude_weight_decay: false
exclude_weight_decay_conf: {}
optim: adam
optim_conf:
    lr: 5.0e-06
    betas:
    - 0.9
    - 0.98
scheduler: tristagelr
scheduler_conf:
    max_steps: 30000
    warmup_ratio: 0.3
    hold_ratio: 0.2
    decay_ratio: 0.5
    init_lr_scale: 0.6
    final_lr_scale: 0.1
init: null
use_preprocessor: true
input_size: null
target_duration: 3.0
lang2utt: dump/raw/train_voxlingua107_lang/lang2utt
lang_num: 107
sample_rate: 16000
num_eval: 10
rir_scp: ''
model: upstream_condition
model_conf:
    lang2vec_conditioning_layers:
    - 32
    - 36
    - 40
    - 44
    apply_intermediate_lang2vec_loss: true
    apply_intermediate_lang2vec_condition: true
    inter_lang2vec_loss_weight: 0.4
    cutoff_gradient_from_backbone: false
    cutoff_gradient_before_condproj: true
    shared_conditioning_proj: true
frontend: s3prl_condition
frontend_conf:
    frontend_conf:
        upstream: hf_wav2vec2_condition
        path_or_url: facebook/mms-1b
    download_dir: ./hub
    multilayer_feature: true
specaug: null
specaug_conf: {}
normalize: utterance_mvn
normalize_conf:
    norm_vars: false
encoder: ecapa_tdnn
encoder_conf:
    model_scale: 8
    ndim: 512
    output_size: 1536
pooling: chn_attn_stat
pooling_conf: {}
projector: rawnet3
projector_conf:
    output_size: 192
encoder_condition: identity
encoder_condition_conf: {}
pooling_condition: chn_attn_stat
pooling_condition_conf: {}
projector_condition: rawnet3
projector_condition_conf: {}
preprocessor: lid
preprocessor_conf:
    fix_duration: false
    sample_rate: 16000
    noise_apply_prob: 0.0
    noise_info:
    -   - 1.0
        - dump/raw/musan_speech.scp
        -   - 4
            - 7
        -   - 13
            - 20
    -   - 1.0
        - dump/raw/musan_noise.scp
        -   - 1
            - 1
        -   - 0
            - 15
    -   - 1.0
        - dump/raw/musan_music.scp
        -   - 1
            - 1
        -   - 5
            - 15
    rir_apply_prob: 0.0
    rir_scp: dump/raw/rirs.scp
    use_lang2vec: true
    lang2vec_type: geo
loss: aamsoftmax_sc_topk_lang2vec
loss_conf:
    margin: 0.5
    scale: 30
    K: 3
    mp: 0.06
    k_top: 5
    lang2vec_dim: 299
    lang2vec_type: geo
    lang2vec_weight: 0.2
required:
- output_dir
version: '202506'
distributed: false
Citation 
@inproceedings{wang2025geolid,
  author={Qingzheng Wang, Hye-jin Shim, Jiancheng Sun, and Shinji Watanabe},
  title={Geolocation-Aware Robust Spoken Language Identification},
  year={2025},
  booktitle={Procedings of ASRU},
}

@inproceedings{watanabe2018espnet,
  author={Shinji Watanabe and Takaaki Hori and Shigeki Karita and Tomoki Hayashi and Jiro Nishitoba and Yuya Unno and Nelson Yalta and Jahn Heymann and Matthew Wiesner and Nanxin Chen and Adithya Renduchintala and Tsubasa Ochiai},
  title={{ESPnet}: End-to-End Speech Processing Toolkit},
  year={2018},
  booktitle={Proceedings of Interspeech},
  pages={2207--2211},
  doi={10.21437/Interspeech.2018-1456},
  url={http://dx.doi.org/10.21437/Interspeech.2018-1456}
}

Runs of espnet geolid_vl107only_shared_trainable on huggingface.co

18
Total runs
0
24-hour runs
2
3-day runs
2
7-day runs
5
30-day runs

More Information About geolid_vl107only_shared_trainable huggingface.co Model

More geolid_vl107only_shared_trainable license Visit here:

https://choosealicense.com/licenses/cc-by-4.0

geolid_vl107only_shared_trainable huggingface.co

geolid_vl107only_shared_trainable huggingface.co is an AI model on huggingface.co that provides geolid_vl107only_shared_trainable's model effect (), which can be used instantly with this espnet geolid_vl107only_shared_trainable model. huggingface.co supports a free trial of the geolid_vl107only_shared_trainable model, and also provides paid use of the geolid_vl107only_shared_trainable. Support call geolid_vl107only_shared_trainable model through api, including Node.js, Python, http.

geolid_vl107only_shared_trainable huggingface.co Url

https://huggingface.co/espnet/geolid_vl107only_shared_trainable

espnet geolid_vl107only_shared_trainable online free

geolid_vl107only_shared_trainable huggingface.co is an online trial and call api platform, which integrates geolid_vl107only_shared_trainable's modeling effects, including api services, and provides a free online trial of geolid_vl107only_shared_trainable, you can try geolid_vl107only_shared_trainable online for free by clicking the link below.

espnet geolid_vl107only_shared_trainable online free url in huggingface.co:

https://huggingface.co/espnet/geolid_vl107only_shared_trainable

geolid_vl107only_shared_trainable install

geolid_vl107only_shared_trainable is an open source model from GitHub that offers a free installation service, and any user can find geolid_vl107only_shared_trainable on GitHub to install. At the same time, huggingface.co provides the effect of geolid_vl107only_shared_trainable install, users can directly use geolid_vl107only_shared_trainable installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

geolid_vl107only_shared_trainable install url in huggingface.co:

https://huggingface.co/espnet/geolid_vl107only_shared_trainable

Url of geolid_vl107only_shared_trainable

geolid_vl107only_shared_trainable huggingface.co Url

Provider of geolid_vl107only_shared_trainable huggingface.co

espnet
ORGANIZATIONS
huggingface.co

Other API from espnet

 huggingface.co
espnet/owsm_ctc_v4_1B

Total runs: 12.1K
Run Growth: 6.9K
Growth Rate: 56.27%
Updated:September 20 2026
huggingface.co
espnet/hubert_dummy

Total runs: 5.5K
Run Growth: 3.7K
Growth Rate: 64.87%
Updated:September 18 2026
huggingface.co
espnet/powsm

Total runs: 183
Run Growth: 124
Growth Rate: 67.76%
Updated:January 22 2026
huggingface.co
espnet/powsm_ctc

Total runs: 116
Run Growth: 64
Growth Rate: 54.70%
Updated:September 20 2026
huggingface.co
espnet/xeus

Total runs: 73
Run Growth: 6
Growth Rate: 8.45%
Updated:June 17 2025
huggingface.co
espnet/owsm_v3.1_ebf

Total runs: 67
Run Growth: -12
Growth Rate: -17.91%
Updated:September 20 2026
huggingface.co
espnet/owsm_v3

Total runs: 21
Run Growth: 15
Growth Rate: 71.43%
Updated:September 20 2026
huggingface.co
espnet/owsm_v1

Total runs: 18
Run Growth: 16
Growth Rate: 88.89%
Updated:September 20 2026