DSP Group® has been a leader in DSP voice algorithms for over two decades. The company's best-in-class algorithms are implemented in ARM processors, DSP processors and other hardware embedded in its products. Supporting current and future requirements for increased functionality and higher quality, the algorithms offer the industry's best cost/performance/power consumption ratio.
DSP Group voice technologies include proprietary innovative speech compression and enhancement, noise reduction, speech enhancement, and echo cancellation algorithms. They also support all industry-standard VoIP and communication voice codecs, as well as communication modems used for telephony and fax signaling.
In the audio domain, DSP Group uses major audio compression technologies such as WMA and MP3.
DSP Group's voice and audio technologies belong to two families covering the full spectrum of digital cordless and VoIP applications.
CleaRplay– Speech and melody playback technologies that improve the clarity and richness of sound
-
Smart melody ringer (SMR) – Low-cost wideband polyphonic melody ringer with up to 15 poly tones , supporting a wide range of instruments; part of DSP Group's DCE chipset family
-
Multiple algorithm synthesis (MAS) – High-quality wideband polyphonic ringer, supporting various sampling rates, boosting functionality, diverse instruments (e.g. drums), reverberation, and special sound effects; part of DSP Group's DCX chipset family
-
Text-to-speech (TTS) – Automatic generation of speech from text in different languages that is mainly used for talking caller ID
-
Flexible bit-rate voice recording and playback – High-quality algorithms powering over 80% of all digital telephone answering machines (DTAMs) worldwide
CleaRtalk- Complementary voice-enhancement and noise-reduction algorithms for improving voice intelligibility
-
Flexible noise reduction (FNR) – Reduction of incoming and outgoing stationary noise applied to both the base station and handset and used for narrowband and wideband (HD voice)
-
Flexible listening enhancement (FLE) single/dual input– Improvement of incoming voice in noisy environments via audio-level adjustments based on acoustic-ambient noise of the listener's environment
-
Enhanced wideband speech synthesis (WBS) – Intelligibility improvement of narrowband speech signals that address humans' wider hearing threshold, resulting in more natural speed synthesis; assists hearing impaired and elderly individuals
-
Enhanced acoustical echo canceller (EAEC) – High-grade, true wideband full-duplex hands-free performance. Improving double-talk performance (meets the strictest CAT-iq 2.x requirements) and Improved echo suppression in narrowband and wideband modes
