Package 'speechmatics'

Delete a job

Description

Delete a job

Usage

sm_delete_job(job_id)

Arguments

job_id

Job ID string.

Value

TRUE, invisibly.

Examples

## Not run: 
sm_delete_job("2om9psu1np")

## End(Not run)

---

Configure channel diarization

Description

Configure channel diarization

Usage

sm_diarize_channel(labels = NULL)

Arguments

labels

Character vector of channel labels (e.g. c("Agent", "Caller")).

Value

A diarization config list.

Examples

sm_diarize_channel()

# label channels
sm_diarize_channel(labels = c("Nic", "Jess"))

---

Configure speaker diarization

Description

Configure speaker diarization

Usage

sm_diarize_speaker(sensitivity = 0.5, prefer_current = FALSE)

Arguments

sensitivity	Numeric 0-1 (default 0.5). Higher values detect more speakers.
prefer_current	Logical. Reduces false switches between similar-sounding speakers.

Value

A diarization config list.

Examples

sm_diarize_speaker()
sm_diarize_speaker(sensitivity = 0.8, prefer_current = TRUE)

---

Get transcript for a completed job

Description

Get transcript for a completed job

Usage

sm_get_transcript(job_id, format = c("txt", "srt", "json-v2"))

Arguments

job_id	Job ID string.
format	Output format: "txt" for plain text (default), "srt" for subtitles, or "json-v2" for structured data with timestamps, speaker labels, and confidence scores.

Value

Transcript as a character string ("txt" and "srt") or a list ("json-v2").

Examples

## Not run: 
sm_get_transcript("2om9psu1np")
sm_get_transcript("2om9psu1np", format = "srt")
sm_get_transcript("2om9psu1np", format = "json-v2")

## End(Not run)

---

List all jobs

Description

List all jobs

Usage

sm_list_jobs()

Value

A data frame of jobs.

Examples

## Not run: 
sm_list_jobs()

## End(Not run)

---

Transcribe an audio file

Description

Submits an audio file to the Speechmatics API, polls until complete, and writes the transcript to a file.

Usage

sm_transcribe(
  input,
  output = NULL,
  config = sm_transcription_config(),
  poll_interval = 5
)

Arguments

input	Path to the input audio file.
output	Path to the output transcript file. If NULL (the default), the output is written to the working directory with the same name as the input file but with a .txt extension.
config	Config from sm_transcription_config().
poll_interval	Seconds between status checks.

Value

The output path, invisibly.

Examples

## Not run: 
audio <- system.file("extdata", "testrecording.mp3", package = "speechmatics")
sm_transcribe(audio)
sm_transcribe(audio, "output.txt")
sm_transcribe(audio, config = sm_transcription_config(quality = "enhanced"))

## End(Not run)

---

Build transcription config

Description

Build transcription config

Usage

sm_transcription_config(
  language = "en",
  quality = c("standard", "enhanced"),
  diarization = NULL
)

Arguments

language	Language code (default "en").
quality	"standard" or "enhanced".
diarization	A diarization object from sm_diarize_speaker() or sm_diarize_channel(), or NULL for no diarization.

Value

A config list suitable for passing to sm_transcribe().

Examples

sm_transcription_config()
sm_transcription_config(language = "fr", quality = "enhanced")
sm_transcription_config(diarization = sm_diarize_speaker())

---