{ "whisper_model": "base", "n_sentences": 20, "voice": "af_heart", "engines": { "ours (Arm CPU, 9 threads)": { "mean_wer": 0.0362, "mean_cer": 0.0308, "perfect": 17, "n": 20, "per_sentence": [ { "reference": "The weather changed suddenly this afternoon.", "transcript": "The weather changed suddenly this afternoon.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/ours_arm_cpu_9_threads_00.wav" }, { "reference": "Could you please pass me that book?", "transcript": "Could you please pass me that book?", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/ours_arm_cpu_9_threads_01.wav" }, { "reference": "Wow, I did not expect to see you here!", "transcript": "Wow, I did not expect to see you here.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/ours_arm_cpu_9_threads_02.wav" }, { "reference": "Our meeting has been moved to three o'clock.", "transcript": "Our meeting has been moved to three o'clock.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/ours_arm_cpu_9_threads_03.wav" }, { "reference": "She walked quietly across the empty room.", "transcript": "She walked quietly across the empty room.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/ours_arm_cpu_9_threads_04.wav" }, { "reference": "How many people are coming to dinner tonight?", "transcript": "How many people are coming to dinner tonight?", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/ours_arm_cpu_9_threads_05.wav" }, { "reference": "The bridge was built over a hundred years ago.", "transcript": "The bridge was built over a hundred years ago.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/ours_arm_cpu_9_threads_06.wav" }, { "reference": "I really appreciate everything you have done.", "transcript": "I really appreciate everything you have done.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/ours_arm_cpu_9_threads_07.wav" }, { "reference": "Turn left at the second traffic light.", "transcript": "Turn left at the second traffic light.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/ours_arm_cpu_9_threads_08.wav" }, { "reference": "This recipe calls for two cups of sugar.", "transcript": "This recipe calls for two cups of shower.", "wer": 0.125, "cer": 0.1026, "audio": "qa/quality_audio/ours_arm_cpu_9_threads_09.wav" }, { "reference": "Are you sure that's the right address?", "transcript": "Are you sure that's the right address?", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/ours_arm_cpu_9_threads_10.wav" }, { "reference": "The engine made a strange noise before it stopped.", "transcript": "The engine made a strange noise before it stopped.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/ours_arm_cpu_9_threads_11.wav" }, { "reference": "He finished the marathon in just under four hours.", "transcript": "He finished the marathon in just under four hours.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/ours_arm_cpu_9_threads_12.wav" }, { "reference": "Please remember to lock the door on your way out.", "transcript": "Please remember to lock the door on your way out.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/ours_arm_cpu_9_threads_13.wav" }, { "reference": "What a beautiful sunset over the mountains.", "transcript": "What a beautiful sunset over the mountains.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/ours_arm_cpu_9_threads_14.wav" }, { "reference": "The report is due by the end of next week.", "transcript": "The record is due by the end of next week.", "wer": 0.1, "cer": 0.0488, "audio": "qa/quality_audio/ours_arm_cpu_9_threads_15.wav" }, { "reference": "I can't believe how fast the year went by.", "transcript": "I can't believe how fast the year went by.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/ours_arm_cpu_9_threads_16.wav" }, { "reference": "Set the oven to three hundred fifty degrees.", "transcript": "Sat the oven to 350 degrees.", "wer": 0.5, "cer": 0.4651, "audio": "qa/quality_audio/ours_arm_cpu_9_threads_17.wav" }, { "reference": "They rebuilt the old library after the fire.", "transcript": "They rebuilt the old library after the fire.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/ours_arm_cpu_9_threads_18.wav" }, { "reference": "Can you hear me clearly on this call?", "transcript": "Can you hear me clearly on this call?", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/ours_arm_cpu_9_threads_19.wav" } ] }, "Kokoro-82M (CPU)": { "mean_wer": 0.0299, "mean_cer": 0.0321, "perfect": 18, "n": 20, "per_sentence": [ { "reference": "The weather changed suddenly this afternoon.", "transcript": "The weather changed suddenly this afternoon.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/kokoro_82m_cpu_00.wav" }, { "reference": "Could you please pass me that book?", "transcript": "Could you please pass me that book?", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/kokoro_82m_cpu_01.wav" }, { "reference": "Wow, I did not expect to see you here!", "transcript": "Wow, I did not expect to see you here.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/kokoro_82m_cpu_02.wav" }, { "reference": "Our meeting has been moved to three o'clock.", "transcript": "Our meeting has been moved to three o'clock.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/kokoro_82m_cpu_03.wav" }, { "reference": "She walked quietly across the empty room.", "transcript": "She walked quietly across the empty room.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/kokoro_82m_cpu_04.wav" }, { "reference": "How many people are coming to dinner tonight?", "transcript": "How many people are coming to dinner tonight?", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/kokoro_82m_cpu_05.wav" }, { "reference": "The bridge was built over a hundred years ago.", "transcript": "The bridge was built over 100 years ago.", "wer": 0.2222, "cer": 0.2, "audio": "qa/quality_audio/kokoro_82m_cpu_06.wav" }, { "reference": "I really appreciate everything you have done.", "transcript": "I really appreciate everything you have done.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/kokoro_82m_cpu_07.wav" }, { "reference": "Turn left at the second traffic light.", "transcript": "Turn left at the second traffic light.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/kokoro_82m_cpu_08.wav" }, { "reference": "This recipe calls for two cups of sugar.", "transcript": "This recipe calls for two cups of sugar.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/kokoro_82m_cpu_09.wav" }, { "reference": "Are you sure that's the right address?", "transcript": "Are you sure that's the right address?", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/kokoro_82m_cpu_10.wav" }, { "reference": "The engine made a strange noise before it stopped.", "transcript": "The engine made a strange noise before it stopped.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/kokoro_82m_cpu_11.wav" }, { "reference": "He finished the marathon in just under four hours.", "transcript": "He finished the marathon in just under four hours.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/kokoro_82m_cpu_12.wav" }, { "reference": "Please remember to lock the door on your way out.", "transcript": "Please remember to lock the door on your way out.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/kokoro_82m_cpu_13.wav" }, { "reference": "What a beautiful sunset over the mountains.", "transcript": "What a beautiful sunset over the mountains.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/kokoro_82m_cpu_14.wav" }, { "reference": "The report is due by the end of next week.", "transcript": "The report is due by the end of next week.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/kokoro_82m_cpu_15.wav" }, { "reference": "I can't believe how fast the year went by.", "transcript": "I can't believe how fast the year went by.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/kokoro_82m_cpu_16.wav" }, { "reference": "Set the oven to three hundred fifty degrees.", "transcript": "Set the oven to 350 degrees.", "wer": 0.375, "cer": 0.4419, "audio": "qa/quality_audio/kokoro_82m_cpu_17.wav" }, { "reference": "They rebuilt the old library after the fire.", "transcript": "They rebuilt the old library after the fire.", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/kokoro_82m_cpu_18.wav" }, { "reference": "Can you hear me clearly on this call?", "transcript": "Can you hear me clearly on this call?", "wer": 0.0, "cer": 0.0, "audio": "qa/quality_audio/kokoro_82m_cpu_19.wav" } ] } } }