speaker-segmentation-eng

This model is a fine-tuned version of openai/whisper-small on the diarizers-community/callhome dataset. It achieves the following results on the evaluation set:

• Loss: 0.4666
• Der: 0.1827
• False Alarm: 0.0590
• Missed Detection: 0.0715
• Confusion: 0.0522

Model description

This segmentation model has been trained on English data (Callhome) using diarizers. It can be loaded with two lines of code:

from diarizers import SegmentationModel

segmentation_model = SegmentationModel().from_pretrained('foduucom/speaker-segmentation-eng')

To use it within a pyannote speaker diarization pipeline, load the pyannote/speaker-diarization-3.1 pipeline, and convert the model to a pyannote compatible format:

from diarizers import SegmentationModel
from pyannote.audio import Pipeline
from datasets import load_dataset
import torch

device = torch.device("cuda:0") if torch.cuda.is_available() else torch.device("cpu")

# load the pre-trained pyannote pipeline
pipeline = Pipeline.from_pretrained("pyannote/speaker-diarization-3.1")
pipeline.to(device)

model = SegmentationModel().from_pretrained("foduucom/speaker-segmentation-eng")
model = model.to_pyannote_model()
pipeline._segmentation.model = model.to(device)

You can now use the pipeline on audio examples:

from datasets import load_dataset
# load dataset example
dataset = load_dataset("diarizers-community/callhome", "eng", split="data")
sample = dataset[0]["audio"]

# pre-process inputs
sample["waveform"] = torch.from_numpy(sample.pop("array")[None, :]).to(device, dtype=model.dtype)
sample["sample_rate"] = sample.pop("sampling_rate")

# perform inference
diarization = pipeline(sample)

# dump the diarization output to disk using RTTM format
with open("audio.rttm", "w") as rttm:
    diarization.write_rttm(rttm)

You can now use the pipeline on single audio examples:

from diarizers import SegmentationModel
from pyannote.audio import Pipeline
from datasets import load_dataset
import torch

device = torch.device("cuda:0") if torch.cuda.is_available() else torch.device("cpu")

# load the pre-trained pyannote pipeline
pipeline = Pipeline.from_pretrained("pyannote/speaker-diarization-3.1")
pipeline.to(device)

model = SegmentationModel().from_pretrained("foduucom/speaker-segmentation-eng")
model = model.to_pyannote_model()
pipeline._segmentation.model = model.to(device)

diarization = pipeline("audio.wav")
with open("audio.rttm", "w") as rttm:
    diarization.write_rttm(rttm)

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.001
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 10

Training results

Framework versions

• Transformers 4.40.1
• Pytorch 2.2.1+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1

Model Card Contact

For inquiries and contributions, please contact us at info@foduu.com.

@ModelCard{
  author    = {Nehul Agrawal and
               Rahul parihar},
  title     = {Speaker Diarization in english language},
  year      = {2024}
}