Grenoble, France –Intuitive and simple user interface drives the growing demand for voice control, either complementing or replacing keyboards, touchscreens and other traditional controls.
The multiplicity of solutions available for Voice Activity Detection combined with absence of vendor-independent benchmark makes it hard for any purchaser to select the best solution. Today, neither unified performances nor benchmark exist to define the quality of a Voice Activity Detector, leaving OEM and Fabless companies using an audio DSP in a tricky situation.
Dolphin Integration fills this benchmarking gap by making publicly available MIWOK™, a vendor-independent benchmark based on objective criteria to assess the performance of any VAD solutions.
For those who are not satisfied by the subjectivity of a mere demonstration, the MIWOK™ benchmark provides to companies embedding, using or developing a VAD to statistically assess and specify the key performances.
What is MIWOK™?
MIWOK™ is an open benchmark, under the Creative Commons ShareAlike license, containing a set of words (representative of language characteristics) and noises (representative of multiple near-field and far-field environments).
What can be measured with MIWOK™?
With MIWOK™ benchmark, the following performances may be measured:
- Detection Latency (DL): it must be short
- Noise detected as Voice (NDV): it must be low
- Voice detected as Voice (VDV): it must be high
On which type of VAD implementation can MIWOK™ be applied?
MIWOK™ is applicable to any VAD implementations:
- Hardware implementation with high-level language models described in Matlab, Verilog, Verilog-A,…
- Software code (C/C++…) for DSP or Application Processor
- Final product with an analog, I2S or PDM interface and featuring an access to the wake-up interrupt signal
How does MIWOK guarantee the representativeness of the complexity of a language?
MIWOK™ benchmark encompasses, for a given language, a set of words that are representative of the first phonemes used. These words have been registered from both mans and women to ensure a realistic spectrum repartition on the audio bandwidth.
MIWOK™ is available in diverse languages (Chinese and English) and both for Near-Field or Far-Field situations.
This allows Dolphin Integration to launch its innovative voice activity detector - WhisperTrigger™ - with factual demonstration of its superior performances in VAD. The WhisperTrigger™ is proposed as a hard block for integration on silicon in either a digital microphone, an application processor or a DSP to wake-up the system without voice data loss.
Such uLP trigger IPs are proposed in numerous processes from 180 nm down to 16 nm, They are ideally suited for triggering key word spotting or voice recognition algorithms thanks to their low detection latency enabling a fast awakening of the voice subsystem.
For further information on MIWOK™ benchmark and WhisperTrigger™, please click here
