Streaming
72 JPY/hr
Streaming recognition with real-time transcript
Up to 37 hours per connection
Pricing
Start for free, scale seamlessly, and pay only for what you use. DolphinVoice's billing model has no monthly fixed fees and offers tiered rates.
Free
- Access to industry-leading speech recognition and synthesis models
- Speech Recognition (Speech-to-Text) Free Trial: 600 minutes/month
- Speech Synthesis (Text-to-Speech) Free Trial: 10k chars/month
- Concurrency Limits: 10 for streaming, 2 for pre-recorded
- Developer docs, email-based support, and resources to help you build
Speech Recognition
Speech-to-Text
※ All prices are in Japanese Yen (tax inclusive).
※ Enterprise customers can take advantage of volume discounts (for over 5,000 hours per month).
| Features | Streaming | Pre-recorded (Standard) | Pre-recorded (VIP) |
|---|---|---|---|
| Basic Specs | |||
Audio Limitation | Up to 37 hours per connection | Audio: 1GB, Video: 2GB, Duration: 5 hours | Audio: 1GB, Duration: 5 hours |
Supported Formats | WAV/PCM/MP3 | WAV/PCM/OPUS/MP3/MP4/M4A/AMR/3GP/AAC | WAV/PCM/OPUS/MP3/AMR/3GP/AAC |
Sampling Rate | 16kHz, 8kHz | 16kHz, 8kHz | 16kHz, 8kHz |
Language Support | Japanese, English, Chinese (supports mixed Chinese-English, mixed Japanese-English) | Japanese, English, Chinese (supports mixed Chinese-English, mixed Japanese-English) | Japanese, English, Chinese (supports mixed Chinese-English, mixed Japanese-English) |
| Pre-recorded Speech-to-Text | |||
Processing Speed | - | 1-hour audio in as fast as 10 minutes | 1-hour audio in as fast as 2 minutes |
Output Formats | - | Script / Subtitle | Script / Subtitle |
Speaker Diarization | - | ||
Speech Rate Calculation | - | ||
| Advanced Features | |||
Inverse Text Normalization (ITN) | |||
Hotwords | |||
Forced Correction | |||
Forbidden Words | |||
Disfluency Detection | |||
Word-level Information | |||
Intermediate Results | |||
Amplitude Gain | |||