Skip to contents

Re-encode Media File with Custom Parameters

prep_recode.Rd

Re-encodes any media file (audio/video) supported by the av package into 16-bit PCM WAV, with optional resampling, channel remixing, and time windowing. Returns the audio data in the same format as av::read_audio_bin.

Usage

prep_recode(
  listOfFiles,
  codec,
  sample_rate = NULL,
  bit_rate = NULL,
  start_time = NULL,
  end_time = NULL,
  channels = NULL,
  verbose = TRUE,
  ...
)

Arguments

listOfFiles

Character vector of file paths to media files

codec

Either "none" (read the file as-is, no re-encoding) or "pcm_s16le" (re-encode to 16-bit PCM WAV). Required. These are the only two re-encoding needs superassp has internally; for anything else call av::av_audio_convert() directly.

sample_rate

Target sample rate in Hz (default: NULL keeps original)

bit_rate

Ignored for "pcm_s16le" (lossless); kept for interface symmetry with av::av_audio_convert().

start_time

Start time in seconds (default: NULL = start of file)

end_time

End time in seconds (default: NULL = end of file)

channels

Number of output channels: 1 (mono), 2 (stereo), or NULL (keep original)

verbose

Logical; show progress messages (default: TRUE)

...

Additional arguments passed to av::av_audio_convert

Value

For single file: Integer vector with audio samples in s32le format (32-bit signed integers), with attributes:

  • channels: Number of audio channels (integer)

  • sample_rate: Sample rate in Hz (integer)

For multiple files: List of integer vectors, one per file

This matches the format returned by av::read_audio_bin().

Details

Re-encoding goes through a temporary WAV file (av::av_audio_convert() followed by av::read_audio_bin()); the temp file is removed on exit. It's useful for:

  • Converting sample rates for analysis

  • Extracting audio from video files

  • Time-windowing large files

  • Remixing channel count across a corpus

References

Ooms J (2024). av: Working with Audio and Video in R. rOpenSci. R package, https://docs.ropensci.org/av/.

FFmpeg Developers (2024). FFmpeg Codecs Documentation. FFmpeg project. https://ffmpeg.org/ffmpeg-codecs.html.

See also

av_audio_convert, read_audio_bin, av_to_asspDataObj

Examples

if (FALSE) { # \dontrun{
# Read as-is
audio <- prep_recode("speech.wav", codec = "none")

# Extract segment from 1-3 seconds
audio_segment <- prep_recode("long.wav",
                              codec = "pcm_s16le",
                              start_time = 1.0,
                              end_time = 3.0)

# Downsample to 16 kHz
audio_16k <- prep_recode("high_res.wav",
                         codec = "pcm_s16le",
                         sample_rate = 16000)

# Convert to mono
audio_mono <- prep_recode("stereo.wav",
                          codec = "pcm_s16le",
                          channels = 1)

# Batch processing
files <- c("file1.mp4", "file2.wav", "file3.flac")
audio_list <- prep_recode(files,
                          codec = "pcm_s16le",
                          sample_rate = 44100,
                          channels = 1)

# Access audio data (same as av::read_audio_bin)
audio <- prep_recode("test.wav", codec = "pcm_s16le")
cat("Channels:", attr(audio, "channels"), "\n")
cat("Sample rate:", attr(audio, "sample_rate"), "\n")
cat("Duration:", length(audio) / attr(audio, "channels") / attr(audio, "sample_rate"), "s\n")
} # }