Why I only get 5 classes of emotion

[Mahaotian1]

According to the official description, the motion2vec_plus model should be able to categorize speech into 9 classes, but why did I actually run it with only 5 classes（angry、happy、neutral、sad and ）?

usage： inference_pipeline = pipeline( task=Tasks.emotion_recognition, model="iic/emotion2vec_plus_base", device = "gpu:1") rec_result = inference_pipeline("/home/data2/tts_data/raw/opensrc/Emotion Speech Dataset/0007/Neutral/0007_000239.wav", output_dir="./cosyvoice_embedding", granularity="utterance", extract_embedding=False)

[LorantSzaboOxit]

emotion2vec_plus_seed/base/large works with 5 classes (angry, happy, neutral, sad, unk)

emotion2_vec_/base/large/base_finetuned works with 9 classes (angry, disgusted, fearful, happy, neutral, other, sad, surprised, unknown)

if i'm right

[ddlBoJack]

You can use this pipeline to get 9-class labels:

from funasr import AutoModel

model = AutoModel(model="iic/emotion2vec_plus_large")

wav_file = f"{model.model_path}/example/test.wav"
rec_result = model.generate(wav_file, output_dir="./outputs", granularity="utterance", extract_embedding=False)
print(rec_result)

[Mahaotian1]

You can use this pipeline to get 9-class labels:

from funasr import AutoModel

model = AutoModel(model="iic/emotion2vec_plus_large")

wav_file = f"{model.model_path}/example/test.wav"
rec_result = model.generate(wav_file, output_dir="./outputs", granularity="utterance", extract_embedding=False)
print(rec_result)

it works，thanks a lot

[itaowei]

the model_revision in pipeline() should be set to master to use the latest model :)

inference_pipeline = pipeline(
    task=Tasks.emotion_recognition,
    model="iic/emotion2vec_plus_large",
    model_revision = "master")

I just met the same problem, hhh