speech_recognition: waveform_path: "C:/Users/baby7/Desktop/multi_modal_model/sample-15s.wav" task: "speech_recognition" output_audio_key: "Transcription" tests: - name: speech_recognition_test waveform_path: "C:/Users/baby7/Desktop/multi_modal_model/sample-15s.wav" expected_output: "Expected transcription"