Abstract: A device for providing speech to a user has a sound wave generator, a sound wave guide for guiding the sound wave to a user's nasal cavity, an airflow generator for producing airflow through the sound wave guide, a pitch controller, and a volume controller. The device causes air and sound waves to pass through the user's mouth so the user can convert the airflow and sound waves to speech. The device can have a nasal cover or nasal prongs to engage the nose. The resulting speech can be controlled to have a substantially constant volume, and can be controlled to have proper tonal inflection. The sound wave generator can provide more than one frequency at a time. The sound wave generator can be a musical instrument. The device is handheld and portable. The device can be used to train a user to control a singing voice.