Module gtts 

Google Translate TTS, driven exactly as gTTS 2.5 drives it.

Text is pre-processed and, past 100 characters, tokenised with gTTS’s rules, then each chunk is posted to the batchexecute RPC and the base64 audio in each answer is concatenated. The rules use regex lookbehind, which the regex crate does not have, so they are written out by hand below — in the same order and with the same alternation priority as the compiled Python pattern.

Enums

GttsError

Constants

ABBREVIATIONS 🔒

ALL_PUNC 🔒

symbols.ALL_PUNC.

DEFAULT_BASE_URL

MAX_CHARS 🔒

RPC 🔒

TONE_MARKS 🔒

USER_AGENT 🔒

Functions

clean_tokens 🔒

compact_ascii 🔒

json.dumps(obj, separators=(",", ":")) — ASCII-escaped.

decode_base64 🔒

is_other_punc 🔒

is_punc_or_space 🔒

_ALL_PUNC_OR_SPACE.match(t) — nothing but punctuation and ASCII whitespace.

minimize 🔒

_minimize(the_string, " ", max_size).

package_rpc 🔒

gTTS._package_rpc(text) → the form body.

pp_abbreviations 🔒

pre_processors.abbreviations: remove the period after each listed abbreviation (case-insensitive, no word boundary — gTTS’s own regex), one abbreviation per pass, each pass judged on that pass’s input.

pp_end_of_line 🔒

pre_processors.end_of_line: drop -\n.

pp_tone_marks 🔒

pre_processors.tone_marks: a space after every tone mark.

pp_word_sub 🔒

pre_processors.word_sub: Esq. → Esquire, case-insensitive.

py_strip 🔒

Python str.strip().

quote 🔒

urllib.parse.quote(s) with the default safe="/".

reserialize_tight 🔒

Remove the spaces Python’s default separators add, outside strings.

synthesize

gTTS(text=, lang=).save(path) → the concatenated MP3 bytes.

text_parts

gTTS._tokenize(text) — the chunks that become requests.

tokenize 🔒

Tokenizer([tone_marks, period_comma, colon, other_punctuation]).run: re.split on the combined, case-insensitive pattern.