{
"asr": " Recognize the speech and give me the transcription.",
"gender_recognition": " What is the gender of the speaker?",
"dialect_identification": " What is the dialect of the speaker?",
"asr_zh": " 请将语音中的内容写下来。",
"summarization": " Could you capture the main points of this audio in a short summary?",
"translation_ae": " Listen to the speech and translate it into English.",
"asr_de": " Hören Sie sich die Rede an und schreiben Sie ihren Inhalt auf.",
"translation_ec": " Listen to the speech and translate it into Chinese.",
"audiocaption": " Please describe the audio.",
"audiocaption_v2": " Please write down what your hear in the audio.",
"QA": " {}",
"gender_QA": " {}",
"phone_recognition": " Provide the phonetic transcription for the speech.",
"speech_query": " Please answer the question in detail.",
"emotion_recognition": " Describe the emotion of the speaker in one word.",
"lyrics_recognition": " Listen to the song and write down its content.",
"audio_speech_description": " Describe the speech and the background audio",
"speaker_verification": " Do you only hear the same person talking? Answer yes or no.",
"fluent_speech_audio": " Describe the background audio and the speech in a fluent sentence.",
"speech_separation": " Please write down what you hear each person says.",
"audio_story_telling": " Based on the audio, write a story in detail. Your story should be highly related to the audio.",
"speech_audio_query": " Please answer the speaker's question in detail based on the background sound.",
"slot_filling": " According to the speech, what is the {}?",
"music_description": " Listen to this music clip and describe the music.",
"translation_en2ja": " Listen to the speech and translate it into Japanese.",
"translation_en2de": " Listen to the speech and translate it into German.",
"speech_audio_coreasoning": " Use your strong reasoning skills to answer the speaker's question in detail based on the background sound.",
"keywords": " Give me only three keywords of the text.",
"speaker_diarization_asr": " Please recognize each speaker and transcribe their speech content."
}