utils_display.py

from dataclasses import dataclass

# These classes are for user facing column names, to avoid having to change them 
# all around the code when a modif is needed 
@dataclass
class ColumnContent:
    name: str
    type: str 

def fields(raw_class):
    return [v for k, v in raw_class.__dict__.items() if k[:2] != "__" and k[-2:] != "__"]

@dataclass(frozen=True)
class AutoEvalColumn: # Auto evals column
    model = ColumnContent("Model", "markdown")
    avg_wer = ColumnContent("Average WER ⬇️", "number")    
    rtf = ColumnContent("RTFx ⬆️️", "number")
    ami_wer = ColumnContent("AMI", "number")
    e22_wer = ColumnContent("Earnings22", "number")
    gs_wer = ColumnContent("Gigaspeech", "number")
    lsc_wer = ColumnContent("LS Clean", "number")
    lso_wer = ColumnContent("LS Other", "number")
    ss_wer = ColumnContent("SPGISpeech", "number")
    tl_wer = ColumnContent("Tedlium", "number")
    vp_wer = ColumnContent("Voxpopuli", "number")

@dataclass(frozen=True)
class MultilingualColumn: # Multilingual benchmark columns
    model = ColumnContent("Model", "markdown")
    avg_multilingual = ColumnContent("Average WER ⬇️", "number")
    rtf = ColumnContent("RTFx ⬆️️", "number")
    de_avg = ColumnContent("🇩🇪 German", "number")
    fr_avg = ColumnContent("🇫🇷 French", "number")
    es_avg = ColumnContent("🇪🇸 Spanish", "number")
    it_avg = ColumnContent("🇮🇹 Italian", "number")
    nl_avg = ColumnContent("🇳🇱 Dutch", "number")
    pl_avg = ColumnContent("🇵🇱 Polish", "number")
    pt_avg = ColumnContent("🇵🇹 Portuguese", "number")
    cs_avg = ColumnContent("🇨🇿 Czech", "number")
    ro_avg = ColumnContent("🇷🇴 Romanian", "number")
    hu_avg = ColumnContent("🇭🇺 Hungarian", "number")

@dataclass(frozen=True)
class LongformColumn: # Long-form ASR benchmark columns
    model = ColumnContent("Model", "markdown")
    avg_longform = ColumnContent("Average WER ⬇️", "number")
    rtf = ColumnContent("RTFx ⬆️️", "number")
    earnings21_wer = ColumnContent("Earnings21", "number")
    mustc_wer = ColumnContent("MustC", "number")


def make_clickable_model(model_name):
    model_name_list = model_name.split("/")
    if model_name_list[0] == "trt-llm":
        link = "https://github.com/NVIDIA/TensorRT-LLM/tree/main/examples/whisper"
    elif model_name_list[0] == "faster-whisper":
        link = "https://github.com/guillaumekln/faster-whisper"
    elif model_name_list[0] == "Whisper.cpp":
        link = "https://github.com/ggerganov/whisper.cpp"
    elif model_name_list[0] == "WhisperKit":
        link = "https://github.com/argmaxinc/WhisperKit"
    elif model_name_list[0] == "WhisperMLX":
        link = "https://huggingface.co/collections/mlx-community/whisper-663256f9964fbb1177db93dc"
    elif model_name_list[0] == "elevenlabs":
        link = "https://elevenlabs.io/speech-to-text"
    elif model_name_list[0] == "openai" and (model_name_list[1] == "whisper-1" or model_name_list[1] == "gpt-4o-transcribe" or model_name_list[1] == "gpt-4o-mini-transcribe"):
        link = "https://platform.openai.com/docs/guides/speech-to-text"
    elif model_name_list[0] == "assemblyai":
        link = "https://www.assemblyai.com/docs"
    elif model_name_list[0] == "revai":
        link = "https://docs.rev.ai/api/asynchronous/get-started/"
    elif model_name_list[0] == "speechmatics":
        link = "https://www.speechmatics.com/"
    elif model_name_list[0] == "ultravox":
        link = "https://huggingface.co/fixie-ai"
    else:
        link = f"https://huggingface.co/{model_name}"
    return f'<a target="_blank" href="{link}" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">{model_name}</a>'

def styled_error(error):
    return f"<p style='color: red; font-size: 20px; text-align: center;'>{error}</p>"

def styled_warning(warn):
    return f"<p style='color: orange; font-size: 20px; text-align: center;'>{warn}</p>"

def styled_message(message):
    return f"<p style='color: green; font-size: 20px; text-align: center;'>{message}</p>"