-
- South Korea / 2016
AI based Sound Signal Processing - Noise Reduction, Voice Extraction, Voice Recognition
- ◾ Price: Negotiable
- ◾ MOQ: Negotiable
AI based Sound Signal Processing - Noise Reduction, Voice Extraction, Voice Recognition
- ◾ Price: Negotiable
- ◾ MOQ: Negotiable
-
Sound Signal Processing, Noise Reduction, Voice Extraction, Voice Recognition, Emotional Inference
-
Required Quantity
-
Brand name-
-
Shipping / Lead TimeNegotiable / Negotiable
-
Keyword
-
Category
Specifications
| Size | - | Weight | - | Stock | 1,000 Pieces |
|---|---|---|---|---|---|
| Country of Origin | South Korea | ||||
| Production method | Available, OBM, ODM, OEM | ||||
| Hs Code | 8471419000 | ||||
Trade Terms
| Payment Terms | D/A, D/P, L/C, T/T | ||||
|---|---|---|---|---|---|
| Price(FOB) | Negotiable | MOQ | Negotiable | ||
| Transportation | Air Transportation, Express, Land Transportation, Negotiation Other, Ocean Shipping | ||||
| Lead time | Negotiable | Shipping time | Negotiable | ||
Company
-
Country / Year Established
-
Membership
- VIP
Business typeManufacturer
-
Verified Certificate
-
1
-
-
1
-
Main Export Markets
-
Canada
-
France
-
Germany
-
Japan
-
U.S.A
-
Product Description
Product Catalog / Brochure / Spec Sheet
1. AI-Noise Reduction Chip in real time
- AI algorithm : Complex Convolution Recurrent Neural Network
- Extract human voice only after all of noises
- Usage : Walkie-Tolkie, Voice recognition in noisy environment
- We have two kinds of chip, Arm Cortex-A for high-end and Arm Cortex-M for low-end.
2. AI-based Sound Separation
- Separation of Music and Voice
3. Voice Recognition in noisy environment
- English, Japanese, Korean
- Reconition rate : more than 98% in SNR=0dB
1. AI-Noise Reduction Chip in real time
- AI algorithm : Complex Convolution Recurrent Neural Network
- Extract human voice only after all of noises
- Usage : Walkie-Tolkie, Voice recognition in noisy environment
- We have two kinds of chip, Arm Cortex-A for high-end and Arm Cortex-M for low-end.
2. AI-based Sound Separation
- Separation of Music and Voice
3. Voice Recognition in noisy environment
- English, Japanese, Korean
- Reconition rate : more than 98% in SNR=0dB
1. For High-end purpose
- Use Arm Cortex-A series 800MHz ~ 1GHz
- Single or dual core
- Main RAM : 32~64 MB
- Flash memory : 16~32MB
2. For Low-end purpose
- Use Arm Cortex-M series 380MHz ~ 700MHz
- Single or dual core
- Main RAM : 4 MB
- Flash memory : 4~8 MB
