Functions — github.com/aarnphm/whispercpp

CodeHub MCPcopy Create free account

hub / github.com/aarnphm/whispercpp / functions

Functions190 in github.com/aarnphm/whispercpp

⨍ Functions 190◇ Types & classes 21 ↳ Endpoints 2

↓ 51 callersMethodget

src/whispercpp/audio.cc:188

↓ 14 callersMethodbuild

src/whispercpp/context.h:180

↓ 10 callersMethodfrom_pretrained

Load a preconverted model from a given model name. Currently it doesn't support custom preconverted models yet. PRs are welcome. Arg

src/whispercpp/__init__.py:81

↓ 9 callersMethodfrom_enum

src/whispercpp/params.cc:7

↓ 8 callersMethodfull_n_segments

Number of generated text segments A segment can be a few words, a sentence, or even a paragraph.

src/whispercpp/context.cc:457

↓ 7 callersMethodfull

Run the entire model: PCM -> log mel spectrogram -> encoder -> decoder -> text Uses the specified decoding strategy to obtain the text. This is usual

src/whispercpp/context.cc:365

↓ 7 callersMethodto_enum

src/whispercpp/context.h:51

↓ 5 callersMethodfrom_file

src/whispercpp/context.cc:59

↓ 5 callersMethodfull_get_segment_text

Get the text of the specified segment.

src/whispercpp/context.cc:498

↓ 5 callersMethodset_context

setters functions

src/whispercpp/context.h:463

↓ 4 callersMethodclear

src/whispercpp/audio.cc:132

↓ 4 callersMethodload_wav_file

src/whispercpp/api_cpp2py_export.cc:3

↓ 4 callersFunctionpreprocess

(file: Path, sample_rate: int = 16000)

tests/export_test.py:24

↓ 4 callersMethodset_strategy

src/whispercpp/context.h:91

↓ 4 callersMethodtranscribe

Transcribe audio from a given numpy array. Args: data (np.ndarray): Audio data as a numpy array. num_proc (int, optio

src/whispercpp/__init__.py:158

↓ 4 callersMethodwith_print_progress

Whether to print progress information Defaults to false

src/whispercpp/context.h:242

↓ 4 callersMethodwith_print_realtime

Print results from within whisper.cpp. Try to use the callback methods instead: [set_new_segment_callback](FullParams::set_new_segment_callback), [set

src/whispercpp/context.h:252

↓ 3 callersMethodstream_transcribe

src/whispercpp/audio.cc:275

↓ 3 callersMethodtranscribe_from_file

Transcribe audio from a given file. This function uses a simple C++ implementation for loading audio file. Currently only WAV files are suppo

src/whispercpp/__init__.py:192

↓ 3 callersMethodwith_beam_size

src/whispercpp/context.h:67

↓ 3 callersMethodwith_language

Set target language. For auto-detection, set this either to 'auto' or nullptr. defaults to 'en'.

src/whispercpp/context.h:341

↓ 3 callersMethodwith_patience

src/whispercpp/context.h:71

↓ 2 callersMethod_load

Load the module and insert it into the parent's globals.

src/whispercpp/utils.py:118

↓ 2 callersFunctioncompile_ext

↓ 2 callersMethodcopy_for_full

src/whispercpp/params.cc:96

↓ 2 callersMethodfrom_params

Load a preconverted model from a given model name and params. Currently it doesn't support custom preconverted models yet. PRs are welcome.

src/whispercpp/__init__.py:121

↓ 2 callersMethodinit_state

src/whispercpp/context.cc:54

↓ 2 callersMethodn_audio_ctx

Get number of audio context

src/whispercpp/context.cc:312

↓ 2 callersMethodreset_timings

src/whispercpp/context.h:554

↓ 2 callersMethodrun

↓ 2 callersMethodset_init_with_state

check if whether context is set with state

src/whispercpp/context.h:467

↓ 2 callersMethodset_state

src/whispercpp/context.h:464

↓ 2 callersFunctionto_timestamp

500 -> 00:05.000 6000 -> 01:00.000

src/whispercpp/audio.cc:256

↓ 2 callersMethodwith_audio_ctx

[EXPERIMENTAL] Speed-up techniques (can reduce the quality of output) Overwrite the audio context size. Default to 0 to use the default value

src/whispercpp/context.h:318

↓ 2 callersMethodwith_best_of

src/whispercpp/context.h:46

↓ 2 callersMethodwith_duration_ms

Set audio duration in milliseconds to decode. defaults to 0 (decode until end of audio)

src/whispercpp/context.h:207

↓ 2 callersMethodwith_max_tokens

[EXPERIMENTAL] Set the maximum tokens per segment. Default to 0 (no limit).

src/whispercpp/context.h:303

↓ 2 callersMethodwith_n_threads

Set the number of threads to use for decoding. Defaults to min(4, std::thread::hardware_concurrency()).

src/whispercpp/context.h:186

↓ 2 callersMethodwith_print_special

Whether to print special tokens (<SOT>, <EOT>, <BEG>) Defaults to false

src/whispercpp/context.h:235

↓ 2 callersMethodwith_print_timestamps

Whether to print timestamps for each text segment when printing realtime Only has an effect if [set_print_realtime](FullParams::set_print_realtime) is

src/whispercpp/context.h:261

↓ 2 callersMethodwith_single_segment

Force single segment output. This may be useful for streaming. Defaults to false

src/whispercpp/context.h:228

↓ 2 callersMethodwith_speed_up

[EXPERIMENTAL] Speed-up techniques (can reduce the quality of output) Speed-up the audio by 2x using Phase Vocoder defaults to false

src/whispercpp/context.h:311

↓ 2 callersMethodwith_temperature_inc

Set temperatur increase. See https://github.com/openai/whisper/blob/f82bc59f5ea234d4b97fb2860842ed38519f7e65/whisper/transcribe.py#L274-L278 Defaults

src/whispercpp/context.h:393

↓ 2 callersMethodwith_translate

Whether to translate to output to language specified under `language` parameter. Defaults to false.

src/whispercpp/context.h:214

↓ 1 callersMethodAudioCapture

src/whispercpp/audio.h:34

↓ 1 callersFunctionExportAudioApi

src/whispercpp/audio.cc:568

↓ 1 callersFunctionExportContextApi

src/whispercpp/context.cc:572

↓ 1 callersFunctionExportParamsApi

src/whispercpp/params.cc:362

↓ 1 callersFunctionExportSamplingStrategiesApi

src/whispercpp/params.cc:219

↓ 1 callersFunctionas_pyarray

src/whispercpp/api_cpp2py_export.h:53

↓ 1 callersMethodcallback

src/whispercpp/audio.cc:154

↓ 1 callersMethoddecode

Run the Whisper decoder to obtain the logits and probabilities for the next token. Make sure to call whisper_encode() first. tokens + n_tokens is the

src/whispercpp/context.cc:189

↓ 1 callersMethodfree

src/whispercpp/context.cc:92

↓ 1 callersMethodfree_state

src/whispercpp/context.cc:87

↓ 1 callersMethodfrom_buffer

src/whispercpp/context.cc:73

↓ 1 callersMethodfull_get_segment_t0

Get the start time of the specified segment.

src/whispercpp/context.cc:478

↓ 1 callersMethodfull_get_segment_t1

Get the end time of the specified segment.

src/whispercpp/context.cc:488

↓ 1 callersMethodfull_get_token_id

Get the token text of the specified token in the specified segment

src/whispercpp/context.cc:542

↓ 1 callersMethodfull_n_tokens

Get numbers of tokens in specified segments.

src/whispercpp/context.cc:514

↓ 1 callersMethodfull_parallel

Split the input audio in chunks and process each chunk separately using whisper_full_with_state() Result is stored in the default state of the contex

src/whispercpp/context.cc:416

↓ 1 callersMethodget

src/whispercpp/context.h:178

↓ 1 callersFunctionget_model

examples/bindings/explore.py:19

↓ 1 callersMethodinit_device

src/whispercpp/audio.cc:32

↓ 1 callersMethodis_multilingual

check if the model is multilingual

src/whispercpp/context.cc:314

↓ 1 callersMethodlang_str_to_id

Returns id of a given language, raise exception if not found

src/whispercpp/context.cc:238

↓ 1 callersMethodlist_available_devices

src/whispercpp/audio.cc:14

↓ 1 callersFunctionmain

CLI entrypoint for demonstrating how to use the API binding directly.

examples/bindings/explore.py:26

↓ 1 callersFunctionmain

(**kwargs: t.Any)

examples/stream/stream.py:10

↓ 1 callersMethodn_vocab

Get number of vocab

src/whispercpp/context.cc:308

↓ 1 callersMethodpause

src/whispercpp/audio.cc:116

↓ 1 callersMethodprint_timings

perf inform and sys info

src/whispercpp/context.h:553

↓ 1 callersMethodresume

src/whispercpp/audio.cc:100

↓ 1 callersFunctionsdl_poll_events

src/whispercpp/audio.cc:554

↓ 1 callersMethodset_new_segment_callback

called for every newly generated text segments Do not use this function unless you know what you are doing. Defaults to None.

src/whispercpp/params.cc:117

↓ 1 callersMethodto_string

src/whispercpp/params.cc:159

↓ 1 callersMethodtoken_to_str

Convert token id to string. Use the vocabulary in provided context

src/whispercpp/context.cc:352

↓ 1 callersMethodtranscribe

examples/bentoml/locustfile.py:10

↓ 1 callersFunctionupdate_submodules

(directory: str)

↓ 1 callersMethodwith_entropy_thold

Set entropy threshold, similar to OpenAI's compression ratio threshold. See https://github.com/openai/whisper/blob/f82bc59f5ea234d4b97fb2860842ed38519

src/whispercpp/context.h:403

↓ 1 callersMethodwith_length_penalty

Set length penalty. See https://github.com/openai/whisper/blob/f82bc59f5ea234d4b97fb2860842ed38519f7e65/whisper/transcribe.py#L267 for more informatio

src/whispercpp/context.h:385

↓ 1 callersMethodwith_logprob_thold

Set logprob_thold. See https://github.com/openai/whisper/blob/f82bc59f5ea234d4b97fb2860842ed38519f7e65/whisper/transcribe.py#L274-L278 for more inform

src/whispercpp/context.h:412

↓ 1 callersMethodwith_max_initial_ts

Set max intial timestamps. See https://github.com/openai/whisper/blob/f82bc59f5ea234d4b97fb2860842ed38519f7e65/whisper/decoding.py#L97 for more inform

src/whispercpp/context.h:376

↓ 1 callersMethodwith_n_max_text_ctx

Set max tokens from past text as prompt for decoder. defaults to 16384

src/whispercpp/context.h:193

↓ 1 callersMethodwith_no_context

Do not use past translation (if any) as initial prompt for the decoder. Defaults to false.

src/whispercpp/context.h:221

↓ 1 callersMethodwith_no_speech_thold

Set no_speech_thold. Currently (as of v1.2.0) not implemented. Defaults to 0.6.

src/whispercpp/context.h:419

↓ 1 callersMethodwith_offset_ms

Set offset in milliseconds to start decoding from. defaults to 0

src/whispercpp/context.h:200

↓ 1 callersMethodwith_prompt_n_tokens

src/whispercpp/context.h:333

↓ 1 callersMethodwith_prompt_tokens

src/whispercpp/context.h:329

↓ 1 callersMethodwith_split_on_word

[EXPERIMENTAL] split on word rather on token (in conjunction with max_len) defaults to false

src/whispercpp/context.h:296

↓ 1 callersMethodwith_suppress_blank

Set suppress_blank. See https://github.com/openai/whisper/blob/f82bc59f5ea234d4b97fb2860842ed38519f7e65/whisper/decoding.py#L89 for more information.

src/whispercpp/context.h:351

↓ 1 callersMethodwith_suppress_non_speech_tokens

Set suppress none speech tokens. See https://github.com/openai/whisper/blob/7858aa9c08d98f75575035ecd6481f462d66ca27/whisper/tokenizer.py#L224-L253 fo

src/whispercpp/context.h:360

↓ 1 callersMethodwith_temperature

Set initial decoding temperature. Defaults to 1.0. See https://ai.stackexchange.com/a/32478

src/whispercpp/context.h:367

↓ 1 callersMethodwith_thold_pt

[EXPERIMENTAL] Set timestamp token probability threshold. Defaults to 0.01

src/whispercpp/context.h:275

↓ 1 callersMethodwith_thold_ptsum

[EXPERIMENTAL] Set timestamp token sum probability threshold. Defaults to 0.01

src/whispercpp/context.h:282

↓ 1 callersMethodwith_token_timestamps

[EXPERIMENTAL] token-level timestamps default to false

src/whispercpp/context.h:268

MethodCallbackAndContext

src/whispercpp/context.h:122

MethodContainer

src/whispercpp/context.h:113

FunctionPYBIND11_MODULE

src/whispercpp/audio.cc:9

FunctionPYBIND11_MODULE

src/whispercpp/api_cpp2py_export.cc:17

src/whispercpp/params.cc:77

next →1–100 of 190, ranked by callers