VOICEPEAK alternative: free, multilingual text to speech
VOICEPEAK is popular Japanese text-to-speech software from AHS. Its narrators sound natural in Japanese, and emotion sliders make it a favourite for videos and games. It is also paid, Japanese-only and licensed per machine.
This guide sums up what VOICEPEAK offers, when it is the better pick, and when SuperClonica, a free offline engine with voices in 31 languages, fits better.
What VOICEPEAK offers
According to AHS, VOICEPEAK runs on a speech engine developed by Dreamtonics. Emotion presets for joy, fun, sadness and anger can be mixed, and volume, speed, pitch, accent and intonation can be adjusted per word. Speech is generated on your PC after online activation, and the software runs on Windows, macOS and Ubuntu. AHS states that English output is not supported.
Price and licence
The commercial 6-narrator set costs ¥23,800 as a download on the AHS site, and one licence covers one machine. Commercial use of the generated audio is allowed in general, while embedding VOICEPEAK in another app for commercial use requires contacting AHS. A trial version with a character limit is available.
When VOICEPEAK is the better choice
If you work only in Japanese and want native-sounding Japanese narrators with fine emotion control, VOICEPEAK is hard to beat. SuperClonica’s catalog voices were mostly made from English and Russian recordings, so in Japanese they keep a trace of that accent.
When SuperClonica fits better
SuperClonica is free to start: the engine and ten stock voices cost nothing and have no character limit, catalog voices cost $2.99 once or are free, and every voice speaks 31 languages, Japanese included. It runs on Windows, macOS and Linux without activation and can be used from scripts, Python or an HTTP server. The studio can also train a voice from your own recording.
Try it
Open any voice in the catalog, choose Japanese or another language, type 30 to 100 characters and listen free, with no account. For whole scripts install the engine with python -m pip install superclonica-cpu and run superclonica --text-file script.txt --voice F2 --lang ja --output narration.wav offline.