pv_porcupine_mcu.h File Reference

#include <stdbool.h>
#include <stdint.h>
#include "picovoice.h"

Include dependency graph for pv_porcupine_mcu.h:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Typedefs
typedef struct pv_porcupine	pv_porcupine_t

Functions
PV_API void	pv_porcupine_delete (pv_porcupine_t *object)
PV_API int32_t	pv_porcupine_frame_length (void)
PV_API pv_status_t	pv_porcupine_get_min_memory_buffer_size (int32_t preliminary_memory_size, void *preliminary_memory_buffer, int32_t num_keywords, const int32_t *keyword_model_sizes, const void *const *keyword_models, int32_t *min_memory_buffer_size)
PV_API pv_status_t	pv_porcupine_init (const char *access_key, int32_t memory_size, void *memory_buffer, int32_t num_keywords, const int32_t *keyword_model_sizes, const void *const *keyword_models, const float *sensitivities, pv_porcupine_t **object)
PV_API pv_status_t	pv_porcupine_process (pv_porcupine_t *object, const int16_t *pcm, int32_t *keyword_index)
const PV_API char *	pv_porcupine_version (void)

Typedef Documentation

pv_porcupine_t

typedef struct pv_porcupine pv_porcupine_t

Forward declaration for Porcupine wake word engine. It detects utterances of given keywords within an incoming stream of audio in real-time. It processes incoming audio in consecutive frames and for each frame emits the detection result. The number of samples per frame can be attained by calling 'pv_porcupine_frame_length()'. The incoming audio needs to have a sample rate equal to 'pv_sample_rate()' and be 16-bit linearly-encoded. Porcupine operates on single-channel audio.

Definition at line 33 of file pv_porcupine_mcu.h.

Function Documentation

pv_porcupine_delete()

PV_API void pv_porcupine_delete

(

pv_porcupine_t *

object

)

Destructor.

Parameters

object

Porcupine object.

pv_porcupine_frame_length()

PV_API int32_t pv_porcupine_frame_length

(

void

)

Getter for number of audio samples per frame.

Returns

Frame length.

pv_porcupine_get_min_memory_buffer_size()

PV_API pv_status_t pv_porcupine_get_min_memory_buffer_size	(	int32_t	preliminary_memory_size,
		void *	preliminary_memory_buffer,
		int32_t	num_keywords,
		const int32_t *	keyword_model_sizes,
		const void *const *	keyword_models,
		int32_t *	min_memory_buffer_size
	)

Computes the minimum required memory buffer size, in bytes, for the given keyword model. A relatively large value for 'preliminary_memory_buffer' is suggested (e.g., 30 kilobytes). Then, 'pv_porcupine_init' can be called optimally passing a memory buffer with the size of 'min_memory_buffer_size'.

Parameters

	preliminary_memory_size	Memory size in bytes.
	preliminary_memory_buffer	Memory needs to be 8-byte aligned.
	num_keywords	Number of keywords to monitor.
	keyword_model_sizes	Size of each keyword model in bytes.
	keyword_models	Keyword models.
[out]	min_memory_buffer_size	minimum required memory buffer size in bytes.

Returns

Status code. Returns 'PV_STATUS_INVALID_ARGUMENT', 'PV_STATUS_INVALID_STATE', or 'PV_STATUS_OUT_OF_MEMORY' on failure.

pv_porcupine_init()

PV_API pv_status_t pv_porcupine_init	(	const char *	access_key,
		int32_t	memory_size,
		void *	memory_buffer,
		int32_t	num_keywords,
		const int32_t *	keyword_model_sizes,
		const void *const *	keyword_models,
		const float *	sensitivities,
		pv_porcupine_t **	object
	)

Constructor.

Parameters

	access_key	AccessKey obtained from Picovoice Console (https://console.picovoice.ai/).
	memory_size	Memory size in bytes.
	memory_buffer	Memory buffer needs to be 8-byte aligned.
	num_keywords	Number of keywords to monitor.
	keyword_model_sizes	Size of each keyword model in bytes.
	keyword_models	Keyword models.
	sensitivities	Sensitivities for detecting keywords. Each value should be a number within [0, 1]. A higher sensitivity results in fewer misses at the cost of increasing the false alarm rate.
[out]	object	Constructed instance of Porcupine.

Returns

Status code. Returns 'PV_STATUS_INVALID_ARGUMENT' or 'PV_STATUS_OUT_OF_MEMORY' on failure.

pv_porcupine_process()

PV_API pv_status_t pv_porcupine_process	(	pv_porcupine_t *	object,
		const int16_t *	pcm,
		int32_t *	keyword_index
	)

Processes a frame of the incoming audio stream and emits the detection result.

Parameters

	object	Porcupine object.
	pcm	A frame of audio samples. The number of samples per frame can be attained by calling 'pv_porcupine_frame_length()'. The incoming audio needs to have a sample rate equal to 'pv_sample_rate()' and be 16-bit linearly-encoded. Porcupine operates on single-channel audio.
[out]	keyword_index	Index of observed keyword at the end of the current frame. Indexing is 0-based and matches the ordering of keyword models provided to 'pv_porcupine_init()'. If no keyword is detected then it is set to -1.

Returns

Status code. Returns 'PV_STATUS_INVALID_ARGUMENT' or 'PV_STATUS_OUT_OF_MEMORY' on failure.

pv_porcupine_version()

const PV_API char* pv_porcupine_version

(

void

)

Getter for version.

Returns

Version.