Skip to content

Latest commit

 

History

History
187 lines (156 loc) · 10.7 KB

README.md

File metadata and controls

187 lines (156 loc) · 10.7 KB
Raw

MR-HuBERT

Pre-trained models

Main models

Model Pretraining Data Model Paper Reference
MR-HuBERT Base (~97M) Librispeech 960 hr download mono_base
MR-HuBERT Base (~321M) Libri-Light 60k hr download mono_large
Multilingual MR-HuBERT Base (~97M) Voxpopuli 100k hr download multi_base
Multilingual MR-HuBERT Large (~321M) Voxpopuli 100k hr download 400k steps or download 600k steps Not in the paper

Abalation models

Model Pretraining Data Model Paper Reference
MR-HuBERT Base (2-4-6 lyrs) Librispeech 960 hr download (B.1)-a
MR-HuBERT Base (5-2-5 lyrs) Librispeech 960 hr download (B.1)-b
MR-HuBERT Base (6-4-2 lyrs) Librispeech 960 hr download (B.1)-c
MR-HuBERT Base (3res 3-2-2-2-3 lyrs) Librispeech 960 hr download (B.2)-a
MR-HuBERT Base (3res 2-2-4-2-2 lyrs) Librispeech 960 hr download (B.2)-b
MR-HuBERT Base (3res 2-2-2-2-2 lyrs) Librispeech 960 hr download (B.2)-c
MR-HuBERT Base (Simple sampling) Librispeech 960 hr download (B.3)-a
MR-HuBERT Base (Single target) Librispeech 960 hr download (B.4)-a
MR-HuBERT Base (Simple Sampling + single target) Librispeech 960 hr download (B.4)-b
MR-HuBERT Base (Mono-resolution 20ms) Librispeech 960 hr download (B.5)-a
MR-HuBERT Base (3-3-3 lyrs) Librispeech 960 hr download (B.6)-a
MR-HuBERT Base (Mono-resolution 20ms, 3-3-3 lyrs) Librispeech 960 hr download (B.6)-b
MR-HuBERT Base (HuBERT 20ms&40ms units) Librispeech 960 hr download (B.7)-a
MR-HuBERT Base (Encodec 50Hz unit) Librispeech 960 hr download (B.7)-b
MR-HuBERT Base (Encodec 50Hz units and 25Hz units) Librispeech 960 hr download (B.7)-c
MR-HuBERT Base (Encodec 50Hz units stream 0&1 ) Librispeech 960 hr download (B.7)-d
MR-HuBERT Large (no audio norm) LibriLight 60k hr download (B.8)-a
MR-HuBERT Large (check paper ) LibriLight 60k hr download (B.8)-b
MR-HuBERT Large (check paper ) LibriLight 60k hr download (B.8)-c
MR-HuBERT Large (check paper ) LibriLight 60k hr download (B.8)-d
MR-HuBERT Large (check paper ) LibriLight 60k hr download (B.8)-e
MR-HuBERT Large (check paper ) LibriLight 60k hr download (B.8)-f
MR-HuBERT Large (check paper ) LibriLight 60k hr download (B.8)-g
MR-HuBERT Large (check paper ) LibriLight 60k hr download (B.8)-h
MR-HuBERT Large (check paper ) LibriLight 60k hr download (B.8)-i
MR-HuBERT Large (check paper ) LibriLight 60k hr download (B.8)-j
Multilingual MR-HuBERT Large (Simple sampling) Voxpopuli 100k hr download Not in paper
MR-HuBERT xLarge (from HuBERT-base label) LibriLight 60k hr download Not in paper
MR-HuBERT xLarge (from HuBERT-large label) LibriLight 60k hr download Not in paper

Load a model

ckpt_path = "/path/to/the/checkpoint.pt"
models, cfg, task = fairseq.checkpoint_utils.load_model_ensemble_and_task([ckpt_path])
model = models[0]

Train a new model

Data preparation

Follow the steps in ./simple_kmeans to create:

  • {train,valid}.tsv waveform list files with length information
/path/to/your/audio/files
file1.wav\t160000
file2.wav\t154600
...
filen.wav\t54362
  • {train,valid}.km frame-aligned pseudo label files (the order is the same as wavefiles in the tsv file).
44 44 44 48 48 962 962 962 962 962 962 962 962 967 967 967 967 967 967 967 967 370 852 370 ... 18 18 745 745
44 44 44 48 48 962 962 962 147 147 147 147 147 147 147 147 147 147 147 147 176 176 271 271 ... 27 27 745 745
...
44 44 44 48 962 962 962 962 962 962 377 377 377 77 77 852 696 694 433 578 578 82 740 622 ... 27 27 745 745
  • dict.km.txt a dummy dictionary (first column is id, the second is dummy one)
0 1
1 1
2 1
...
999 1

The label_rate is the same as the feature frame rate used for clustering, which is 100Hz for MFCC features and 50Hz for HuBERT features by default.

Pre-train a MR-HuBERT model

Suppose {train,valid}.tsv are saved at /path/to/data, {train,valid}.km are saved at /path/to/labels, and the label rate is 100Hz.

To train a base model (12 layer transformer), run:

$ python fairseq_cli/hydra_train.py \
  --config-dir /path/to/fairseq-py/examples/mr_hubert/config/pretrain \
  --config-name mrhubert_base_librispeech \
  task.data=/path/to/data task.label_dir=/path/to/labels \
  task.labels='["km"]' model.label_rate=100 \
  task.label_rate_ratios='[1, 2]' \

Please see sample pre-training scripts train.sh for an example script.

Fine-tune a MR-HuBERT model with a CTC loss

Suppose {train,valid}.tsv are saved at /path/to/data, and their corresponding character transcripts {train,valid}.ltr are saved at /path/to/trans. A typical ltr file is with the same order of tsv waveform files as

HOW | ARE | YOU
...
THANK | YOU

To fine-tune a pre-trained MR-HuBERT model at /path/to/checkpoint, run

$ python fairseq_cli/hydra_train.py \
  --config-dir /path/to/fairseq-py/examples/mr_hubert/config/finetune \
  --config-name base_10h \
  task.data=/path/to/data task.label_dir=/path/to/trans \
  model.w2v_path=/path/to/checkpoint

Please see sample fine-tuning scripts finetune.sh for an example script.

Decode a MR-HuBERT model

Suppose the test.tsv and test.ltr are the waveform list and transcripts of the split to be decoded, saved at /path/to/data, and the fine-tuned model is saved at /path/to/checkpoint.

We support three decoding modes:

  • Viterbi decoding: greedy decoding without a language model
  • KenLM decoding: decoding with an arpa-format KenLM n-gram language model
  • Fairseq-LM deocding: decoding with a Fairseq neural language model (not fully tested)

Viterbi decoding

task.normalize needs to be consistent with the value used during fine-tuning. Decoding results will be saved at /path/to/experiment/directory/decode/viterbi/test.

$ python examples/speech_recognition/new/infer.py \
  --config-dir /path/to/fairseq-py/examples/mr_hubert/config/decode \
  --config-name infer \
  task.data=/path/to/data \
  task.normalize=[true|false] \
  decoding.exp_dir=/path/to/experiment/directory \
  common_eval.path=/path/to/checkpoint
  dataset.gen_subset=test \

KenLM / Fairseq-LM decoding

Suppose the pronunciation lexicon and the n-gram LM are saved at /path/to/lexicon and /path/to/arpa, respectively. Decoding results will be saved at /path/to/experiment/directory/decode/kenlm/test.

$ python examples/speech_recognition/new/infer.py \
  --config-dir /path/to/fairseq-py/examples/mr_hubert/config/decode \
  --config-name infer_lm \
  task.data=/path/to/data \
  task.normalize=[true|false] \
  decoding.exp_dir=/path/to/experiment/directory \
  common_eval.path=/path/to/checkpoint
  dataset.gen_subset=test \
  decoding.decoder.lexicon=/path/to/lexicon \
  decoding.decoder.lmpath=/path/to/arpa

The command above uses the default decoding hyperparameter, which can be found in examples/speech_recognition/hydra/decoder.py. These parameters can be configured from the command line. For example, to search with a beam size of 500, we can append the command above with decoding.decoder.beam=500. Important parameters include:

  • decoding.decoder.beam
  • decoding.decoder.beamthreshold
  • decoding.decoder.lmweight
  • decoding.decoder.wordscore
  • decoding.decoder.silweight

To decode with a Fairseq LM, you may check the usage examples in wav2vec2 or hubert examples.

Please see sample decoding scripts decode.sh for an example script.