Japanese Scripted Monologue, generic
Start training, testing or fine-tuning your speech models with 599 hours of Japanese recorded on device. This dataset is perfect for those who are looking for short, high-quality recordings of with extremely accurate transcriptions for their ASR models. Recordings are saved .wav files with a sample rate of 16000 or 48000 and a bit depth of 16 bit.
Start training, testing or fine-tuning your speech models with 599 hours of Japanese recorded on device. This dataset is perfect for those who are looking for short, high-quality recordings of with extremely accurate transcriptions for their ASR models. Recordings are saved .wav files with a sample rate of 16000 or 48000 and a bit depth of 16 bit.
Start training, testing or fine-tuning your speech models with 599 hours of Japanese recorded on device. This dataset is perfect for those who are looking for short, high-quality recordings of with extremely accurate transcriptions for their ASR models. Recordings are saved .wav files with a sample rate of 16000 or 48000 and a bit depth of 16 bit.
Start training, testing or fine-tuning your speech models with 599 hours of Japanese recorded on device. This dataset is perfect for those who are looking for short, high-quality recordings of with extremely accurate transcriptions for their ASR models. Recordings are saved .wav files with a sample rate of 16000 or 48000 and a bit depth of 16 bit.
Dataset specs
Type
Audio
Sound quality
8kHz, 16 bit per channel
Region/Locale
JA, ja-JP
Amount
599 hours
Leverage
Advance AI's understanding and generation of natural Japanese speech.
Empower your technologies with precise, controlled speech signals, designed to build accurate and reliable Japanese language understanding at scale.
Use cases
Speech Recognition and Analysis
Natural Language Processing and Understanding
Keyword Spotting and Voice Command Recognition
Do you need a specific dataset?
We understand the uniqueness of every project. That's why we offer customizable dataset solutions to match your specific requirements.
Dataset specs
Type
Audio
Sound quality
8kHz, 16 bit per channel
Region/Locale
JA, ja-JP
Amount
599 hours