webrtc_vad.h (3099B)
1 /* 2 * Copyright (c) 2012 The WebRTC project authors. All Rights Reserved. 3 * 4 * Use of this source code is governed by a BSD-style license 5 * that can be found in the LICENSE file in the root of the source 6 * tree. An additional intellectual property rights grant can be found 7 * in the file PATENTS. All contributing project authors may 8 * be found in the AUTHORS file in the root of the source tree. 9 */ 10 11 /* 12 * This header file includes the VAD API calls. Specific function calls are 13 * given below. 14 */ 15 16 #ifndef COMMON_AUDIO_VAD_INCLUDE_WEBRTC_VAD_H_ // NOLINT 17 #define COMMON_AUDIO_VAD_INCLUDE_WEBRTC_VAD_H_ 18 19 #include <stddef.h> 20 #include <stdint.h> 21 22 typedef struct WebRtcVadInst VadInst; 23 24 #ifdef __cplusplus 25 extern "C" { 26 #endif 27 28 // Creates an instance to the VAD structure. 29 VadInst* WebRtcVad_Create(void); 30 31 // Frees the dynamic memory of a specified VAD instance. 32 // 33 // - handle [i] : Pointer to VAD instance that should be freed. 34 void WebRtcVad_Free(VadInst* handle); 35 36 // Initializes a VAD instance. 37 // 38 // - handle [i/o] : Instance that should be initialized. 39 // 40 // returns : 0 - (OK), 41 // -1 - (null pointer or Default mode could not be set). 42 int WebRtcVad_Init(VadInst* handle); 43 44 // Sets the VAD operating mode. A more aggressive (higher mode) VAD is more 45 // restrictive in reporting speech. Put in other words the probability of being 46 // speech when the VAD returns 1 is increased with increasing mode. As a 47 // consequence also the missed detection rate goes up. 48 // 49 // - handle [i/o] : VAD instance. 50 // - mode [i] : Aggressiveness mode (0, 1, 2, or 3). 51 // 52 // returns : 0 - (OK), 53 // -1 - (null pointer, mode could not be set or the VAD instance 54 // has not been initialized). 55 int WebRtcVad_set_mode(VadInst* handle, int mode); 56 57 // Calculates a VAD decision for the `audio_frame`. For valid sampling rates 58 // frame lengths, see the description of WebRtcVad_ValidRatesAndFrameLengths(). 59 // 60 // - handle [i/o] : VAD Instance. Needs to be initialized by 61 // WebRtcVad_Init() before call. 62 // - fs [i] : Sampling frequency (Hz): 8000, 16000, or 32000 63 // - audio_frame [i] : Audio frame buffer. 64 // - frame_length [i] : Length of audio frame buffer in number of samples. 65 // 66 // returns : 1 - (Active Voice), 67 // 0 - (Non-active Voice), 68 // -1 - (Error) 69 int WebRtcVad_Process(VadInst* handle, 70 int fs, 71 const int16_t* audio_frame, 72 size_t frame_length); 73 74 // Checks for valid combinations of `rate` and `frame_length`. We support 10, 75 // 20 and 30 ms frames and the rates 8000, 16000 and 32000 Hz. 76 // 77 // - rate [i] : Sampling frequency (Hz). 78 // - frame_length [i] : Speech frame buffer length in number of samples. 79 // 80 // returns : 0 - (valid combination), -1 - (invalid combination) 81 int WebRtcVad_ValidRateAndFrameLength(int rate, size_t frame_length); 82 83 #ifdef __cplusplus 84 } 85 #endif 86 87 #endif // COMMON_AUDIO_VAD_INCLUDE_WEBRTC_VAD_H_ // NOLINT