CosyVoice/tools/extract_embedding.py

#!/usr/bin/env python3
# Copyright (c) 2024 Alibaba Inc (authors: Xiang Lyu)
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#   http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import argparse
import torch
import torchaudio
from tqdm import tqdm
import onnxruntime
import torchaudio.compliance.kaldi as kaldi
from queue import Queue, Empty
from threading import Thread


class ExtractEmbedding:
    def __init__(self, model_path: str, queue: Queue, out_queue: Queue):
        self.model_path = model_path
        self.queue = queue
        self.out_queue = out_queue
        self.is_run = True

    def run(self):
        self.consumer_thread = Thread(target=self.consumer)
        self.consumer_thread.start()

    def stop(self):
        self.is_run = False
        self.consumer_thread.join()

    def consumer(self):
        option = onnxruntime.SessionOptions()
        option.graph_optimization_level = (
            onnxruntime.GraphOptimizationLevel.ORT_ENABLE_ALL
        )
        option.intra_op_num_threads = 1
        providers = ["CPUExecutionProvider"]
        ort_session = onnxruntime.InferenceSession(
            self.model_path, sess_options=option, providers=providers
        )

        while self.is_run:
            try:
                utt, wav_file = self.queue.get(timeout=1)

                audio, sample_rate = torchaudio.load(wav_file)
                if sample_rate != 16000:
                    audio = torchaudio.transforms.Resample(
                        orig_freq=sample_rate, new_freq=16000
                    )(audio)
                feat = kaldi.fbank(
                    audio, num_mel_bins=80, dither=0, sample_frequency=16000
                )
                feat = feat - feat.mean(dim=0, keepdim=True)
                embedding = (
                    ort_session.run(
                        None,
                        {
                            ort_session.get_inputs()[0]
                            .name: feat.unsqueeze(dim=0)
                            .cpu()
                            .numpy()
                        },
                    )[0]
                    .flatten()
                    .tolist()
                )
                self.out_queue.put((utt, embedding))
            except Empty:
                self.is_run = False
                break


def main(args):
    utt2wav, utt2spk = {}, {}
    with open("{}/wav.scp".format(args.dir)) as f:
        for l in f:
            l = l.replace("\n", "").split()
            utt2wav[l[0]] = l[1]
    with open("{}/utt2spk".format(args.dir)) as f:
        for l in f:
            l = l.replace("\n", "").split()
            utt2spk[l[0]] = l[1]

    input_queue = Queue()
    output_queue = Queue()
    consumers = [
        ExtractEmbedding(args.onnx_path, input_queue, output_queue)
        for _ in range(args.num_thread)
    ]

    utt2embedding, spk2embedding = {}, {}
    for utt in tqdm(utt2wav.keys(), desc="Load data"):
        input_queue.put((utt, utt2wav[utt]))

    for c in consumers:
        c.run()

    with tqdm(desc="Process data: ", total=len(utt2wav)) as pbar:
        while any([c.is_run for c in consumers]):
            try:
                utt, embedding = output_queue.get(timeout=1)
                utt2embedding[utt] = embedding
                spk = utt2spk[utt]
                if spk not in spk2embedding:
                    spk2embedding[spk] = []
                spk2embedding[spk].append(embedding)
                pbar.update(1)
            except Empty:
                continue

    for k, v in spk2embedding.items():
        spk2embedding[k] = torch.tensor(v).mean(dim=0).tolist()

    torch.save(utt2embedding, "{}/utt2embedding.pt".format(args.dir))
    torch.save(spk2embedding, "{}/spk2embedding.pt".format(args.dir))


if __name__ == "__main__":
    parser = argparse.ArgumentParser()
    parser.add_argument("--dir", type=str)
    parser.add_argument("--onnx_path", type=str)
    parser.add_argument("--num_thread", type=int, default=8)
    args = parser.parse_args()
    main(args)