Enrollment class. More...
#include <snowboy-detect.h>
Public Member Functions | |
| SnowboyPersonalEnroll (const std::string &resource_filename, const std::string &model_filename) | |
| Default constructor. More... | |
| int | RunEnrollment (const std::string &data) |
| Runs the hotword enrollment. More... | |
| int | RunEnrollment (const float *const data, const int array_length) |
| Runs enrollment on float samples. More... | |
| int | RunEnrollment (const int16_t *const data, const int array_length) |
| Runs enrollment on int16_t samples. More... | |
| int | RunEnrollment (const int32_t *const data, const int array_length) |
| Runs enrollment on int32_t samples. More... | |
| bool | Reset () |
| Resets the enrollment. More... | |
| int | GetNumTemplates () const |
| Returns the minimum number of recordings required for the new hotword. More... | |
| int | SampleRate () const |
| Returns the expected sample rate for audio provided to RunDetection(). More... | |
| int | NumChannels () const |
| Returns the expected number of channels for audio provided to RunDetection(). More... | |
| int | BitsPerSample () const |
| Returns the expected number of bits for audio provided to RunDetection(). More... | |
| ~SnowboyPersonalEnroll () | |
| Destructor. | |
Friends | |
| class | testing::Inspector |
Detailed Description
Enrollment class.
Class that allows enrolling new personal hotwords.
Constructor & Destructor Documentation
◆ SnowboyPersonalEnroll()
| snowboy::SnowboyPersonalEnroll::SnowboyPersonalEnroll | ( | const std::string & | resource_filename, |
| const std::string & | model_filename | ||
| ) |
Default constructor.
Constructor that takes a resource file and a hotword model filename. The finished model will be written to the specified file.
- Parameters
-
[in] resource_filename Filename of resource file. [in] model_filename Filename for the generated hotword model.
Member Function Documentation
◆ BitsPerSample()
| int snowboy::SnowboyPersonalEnroll::BitsPerSample | ( | ) | const |
Returns the expected number of bits for audio provided to RunDetection().
- Returns
- The expected number of bits.
◆ GetNumTemplates()
| int snowboy::SnowboyPersonalEnroll::GetNumTemplates | ( | ) | const |
Returns the minimum number of recordings required for the new hotword.
- Returns
- Number of needed recordings.
◆ NumChannels()
| int snowboy::SnowboyPersonalEnroll::NumChannels | ( | ) | const |
Returns the expected number of channels for audio provided to RunDetection().
- Returns
- The expected number of channels.
◆ Reset()
| bool snowboy::SnowboyPersonalEnroll::Reset | ( | ) |
Resets the enrollment.
After reset the pipeline will behave identical to a freshly constructed instance.
- Returns
- true on success
false on error
◆ RunEnrollment() [1/4]
| int snowboy::SnowboyPersonalEnroll::RunEnrollment | ( | const float *const | data, |
| const int | array_length | ||
| ) |
Runs enrollment on float samples.
If NumChannels() > 1, e.g., NumChannels() == 2, then the array is as follows:
d1c1, d1c2, d2c1, d2c2, d3c1, d3c2, ..., dNc1, dNc2
where d1c1 means data point 1 of channel 1.
- Parameters
-
[in] data Recording of the hotword to be enrolled. [in] array_length Length of the data array in elements.
- Returns
- See RunEnrollment(const std::string&) for details
◆ RunEnrollment() [2/4]
| int snowboy::SnowboyPersonalEnroll::RunEnrollment | ( | const int16_t *const | data, |
| const int | array_length | ||
| ) |
Runs enrollment on int16_t samples.
If NumChannels() > 1, e.g., NumChannels() == 2, then the array is as follows:
d1c1, d1c2, d2c1, d2c2, d3c1, d3c2, ..., dNc1, dNc2
where d1c1 means data point 1 of channel 1.
- Parameters
-
[in] data Recording of the hotword to be enrolled. [in] array_length Length of the data array in elements.
- Returns
- See RunEnrollment(const std::string&) for details
◆ RunEnrollment() [3/4]
| int snowboy::SnowboyPersonalEnroll::RunEnrollment | ( | const int32_t *const | data, |
| const int | array_length | ||
| ) |
Runs enrollment on int32_t samples.
If NumChannels() > 1, e.g., NumChannels() == 2, then the array is as follows:
d1c1, d1c2, d2c1, d2c2, d3c1, d3c2, ..., dNc1, dNc2
where d1c1 means data point 1 of channel 1.
- Parameters
-
[in] data Recording of the hotword to be enrolled. [in] array_length Length of the data array in elements.
- Returns
- See RunEnrollment(const std::string&) for details
◆ RunEnrollment() [4/4]
| int snowboy::SnowboyPersonalEnroll::RunEnrollment | ( | const std::string & | data | ) |
Runs the hotword enrollment.
Supported audio format is WAVE (with linear PCM, 8-bits unsigned integer, 16-bits signed integer or 32-bits signed integer). See SampleRate(), NumChannels() and BitsPerSample() for the required sampling rate, number of channels and bits per sample values. You are supposed to provide a full recording of the hotword each time you call RunEnrollment().
- Returns
Code Info -1 Error. 0 No event. 1 Hotword is too long 2 Hotword is too short
- Parameters
-
[in] data Recording of the hotword to enroll.
◆ SampleRate()
| int snowboy::SnowboyPersonalEnroll::SampleRate | ( | ) | const |
Returns the expected sample rate for audio provided to RunDetection().
- Returns
- The expected samplerate.
The documentation for this class was generated from the following files:
- lib/snowboy-detect.h
- lib/snowboy-detect.cpp
