speechtotext.model.assemblyAIWrapper.AssemblyAIAPIWrapper

class AssemblyAIAPIWrapper(model_version)[source]

Bases: ModelWrapper

Wrapper for assemblyAi API. ASSEMBLY_AI_API_KEY needs to be in the ‘.env’ file in current directory.

Wrapper for assemblyAi model.

Parameters:: model_version (assemblyAiAPIVersion) – Model version of assemblyAi STT API to use.

Methods

`benchmark_n_samples`	Benchmark n samples with model.
`benchmark_sample`	Benchmark sample with model.
`benchmark_samples`	Benchmark samples with model.
`convert_sample`	Convert sample to correct format.
`get_model`	Get model.
`get_transcript_of_file`

Attributes

`API_URL`	Connection url for the API.
`LANGUAGE_CODE`	Code for the language to transcribe.
`PATH_OF_TEMP_CONVERTED_AUDIO_FILE`	path to temp file that will be created to convert the audio files to an accepted audio format.
`POLLING_ENDPOINT`	Polling endpoint url.
`TIME_SLEEP`	Time to sleep after each polling.
`TRANSCRIPT_ENDPOINT`	Transcribe endpoint url.
`UPLOAD_ENDPOINT`	Upload endpoint url.

API_URL: str = 'https://api.assemblyai.com/v2/upload'

Connection url for the API.

Type:: str

LANGUAGE_CODE: str = 'nl'

Code for the language to transcribe.

See supported languages for assemblyAi

Type:: str

PATH_OF_TEMP_CONVERTED_AUDIO_FILE: str = 'converted_audio_file.wav'

path to temp file that will be created to convert the audio files to an accepted audio format.

Type:: PATH_OF_TEMP_CONVERTED_AUDIO_FILE

POLLING_ENDPOINT: str = 'https://api.assemblyai.com/v2/transcript/'

Polling endpoint url.

Type:: str

TIME_SLEEP: int = 3

Time to sleep after each polling.

Type:: int

TRANSCRIPT_ENDPOINT: str = 'https://api.assemblyai.com/v2/transcript'

Transcribe endpoint url.

Type:: str

UPLOAD_ENDPOINT: str = 'https://api.assemblyai.com/v2/upload'

Upload endpoint url.

Type:: str

_append_error(samples, audio_id, error)

Append error to model_errors.

Parameters:

samples (SampleDataset) – Dataset of audio.
id (str) – Id of failed sample.
error (str) – Error message.

_benchmark_sample_with_time(dataset, audio_id, with_cleaning=True)

Benchmark sample for model with timer.

Parameters:

dataset (Dataset) – Dataset of audio.
id (str) – Id of audio file.
with_cleaning (bool, optional) – Set True to clean transcripts. Defaults to True.

Returns:

Metrics of the transcript.

Return type:

Metrics

_clean_output(paragraphs)[source]

Transcript list to 1 string.

Parameters:: paragraphs (list) – Transcript list.
Returns:: Transcript.
Return type:: str

_get_paragraphs(polling_endpoint, header)[source]

Get results from polling endpoint.

Parameters:

polling_endpoint (str) – Polling endpoint.
header (dict) – Http header.

Returns:

Transcript of audio file in list.

Return type:

list

_make_polling_endpoint(transcript_response)[source]

Make polling endoint.

Parameters:: transcript_response (dict) – Transcript response.
Returns:: Polling endpoint.
Return type:: str

_read_file_with_chunck_size(audio_file_name, chunk_size=5242880)[source]

Read data from file.

Parameters:

audio_file_name (str) – Path to audio file.
chunk_size (int, optional) – Size of chunk. Defaults to 5242880.

Yields:

generator – Audio file.

_request_transcript(upload_url, header)[source]

Request transcript.

Parameters:

upload_url (dict) – Url for request.
header (dict) – Http header.

Returns:

Response of request.

Return type:

dict

_upload_file(audio_file_name, header)[source]

Upload file.

Parameters:

audio_file_name (str) – Path to audio file.
header (dict) – Http header.

Returns:

Url for request.

Return type:

dict

_wait_for_completion(polling_endpoint, header)[source]

Wait for the translation to be done.

Parameters:

polling_endpoint (str) – Polling endpoint.
header (dict) – Http header.

benchmark_n_samples(dataset, number_of_samples, with_cleaning=True)

Benchmark n samples with model.

Parameters:

dataset (Dataset) – Dataset of audio.
number_of_samples (int) – Number of random samples to benchmerk.
with_cleaning (bool, optional) – Set True to clean transcripts. Defaults to True.

Returns:

List of metrics for each sample.

Return type:

list

benchmark_sample(dataset, audio_id, with_cleaning=True)

Benchmark sample with model.

Parameters:

dataset (Dataset) – Dataset of audio.
id (str) – Id of audio file.
with_cleaning (bool, optional) – Set True to clean transcripts. Defaults to True.

Returns:

Metrics of the transcript.

Return type:

Metrics

benchmark_samples(samples, with_cleaning=True)

Benchmark samples with model.

Parameters:

dataset (Dataset) – Dataset of audio.
number_of_samples (int) – Number of random samples to benchmerk.
with_cleaning (bool, optional) – Set True to clean transcripts. Defaults to True.

Returns:

List of metrics for each sample.

Return type:

list

convert_sample(path_to_sample)

Convert sample to correct format.

Parameters:

path_to_sample (str) – Path to sample.
override (bool, optional) – Override original file. Defaults to False.

Returns:

Path to converted sample.

Return type:

str

get_model()[source]: Get model.