Falcon Speaker Diarization
C Quick Start

Platforms

• Linux (x86_64)
• macOS (x86_64, arm64)
• Windows (x86_64, arm64)
• Raspberry Pi (3, 4, 5)

Requirements

• C99-compatible compiler
• CMake (3.13+)
• For Windows Only: MinGW is required to build the demo

Picovoice Account & AccessKey

Signup or Login to Picovoice Console to get your AccessKey. Make sure to keep your AccessKey secret.

Quick Start

Setup

1. Clone the repository:

git clone --recurse-submodules https://github.com/Picovoice/falcon.git

Usage

1. Include the public header files (picovoice.h and pv_falcon.h).
2. Link the project to an appropriate precompiled library for the target platform and load it.
3. Use the default model file.
4. Construct the Falcon Speaker Diarization object:

static const char* ACCESS_KEY = "${ACCESS_KEY}";

const char *model_file_path = "${MODEL_FILE_PATH}";
const char *device = "best";

pv_falcon_t *falcon;

const pv_status_t status = pv_falcon_init(
    ACCESS_KEY
    model_file_path,
    device,
    &falcon);

if (status != PV_STATUS_SUCCESS) {
    // error handling logic
}

5. Pass in an audio path to the pv_falcon_process_file function:

static const char* audio_path = "${AUDIO_FILE_PATH}";

int32_t num_segments = 0;
pv_segment_t *segments = NULL;
status = pv_falcon_process_file(falcon, audio_path, &num_segments, &segments);
if (status != PV_STATUS_SUCCESS) {
  // error handling logic
}
for (int32_t i = 0; i < num_segments; i++) {
    pv_segment_t *segment = &segments[i];
    fprintf(
            stdout,
            "Speaker: %d -> Start: %5.2f, End: %5.2f\n",
            segment->speaker_tag,
            segment->start_sec,
            segment->end_sec);
}

pv_falcon_segments_delete(segments); // make sure to free segments result

6. Release resources explicitly when done with Falcon Speaker Diarization:

pv_falcon_delete(falcon);

Segments

Falcon Speaker Diarization returns an array of segments. Each segment has the following properties:

• Start Time: Indicates when the segment started. Value is in seconds.
• End Time: Indicates when the segment ended. Value is in seconds.
• Speaker Tag: A non-negative integer identifying unique speakers.

Demo

For the Falcon Speaker Diarization C SDK, we offer demo applications that demonstrate how to use the Speaker Diarization engine on audio files.

Setup

1. Clone the Falcon Speaker Diarization repository from GitHub using HTTPS:

git clone --recurse-submodules https://github.com/Picovoice/falcon.git

2. Build the demo:

cd falcon
cmake -S demo/c/. -B demo/c/build
cmake --build demo/c/build --target falcon_demo

Usage

To see the usage options for the demo:

./demo/c/build/falcon_demo

Run the command corresponding to your platform from the root of the repository:

./demo/c/build/falcon_demo \
-a ${ACCESS_KEY} \
-m ${MODEL_FILE_PATH} \
-l lib/${PLATFORM}/${ARCH}/libpv_falcon.so \
${AUDIO_PATH1} ${AUDIO_PATH2} ...

For more information on our Falcon Speaker Diarization demos for C, head over to our GitHub repository.

Resources