Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

整理: 簡易 docstring と単純変数名 #903

Merged
merged 2 commits into from
Dec 19, 2023
Merged

整理: 簡易 docstring と単純変数名 #903

Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
cc3b349
Refactor: 簡易docstring と単純変数名
tarepan Dec 18, 2023
8a93d79
Merge branch 'master' into refactor/apply_simple
Hiroshiba Dec 19, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
121 changes: 17 additions & 104 deletions voicevox_engine/tts_pipeline/tts_engine.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -123,39 +123,15 @@ def generate_silence_mora(length: float) -> Mora:


def apply_prepost_silence(moras: list[Mora], query: AudioQuery) -> list[Mora]:
"""
前後無音（`prePhonemeLength` & `postPhonemeLength`）の適用
Parameters
----------
moras : List[Mora]
モーラ時系列
query : AudioQuery
音声合成用のクエリ
Returns
-------
moras : List[Mora]
前後無音が付加されたモーラ時系列
"""
"""モーラ系列へ音声合成用のクエリがもつ前後無音（`prePhonemeLength` & `postPhonemeLength`）を付加する"""
pre_silence_moras = [generate_silence_mora(query.prePhonemeLength)]
post_silence_moras = [generate_silence_mora(query.postPhonemeLength)]
moras = pre_silence_moras + moras + post_silence_moras
return moras


def apply_speed_scale(moras: list[Mora], query: AudioQuery) -> list[Mora]:
"""
話速スケール（`speedScale`）の適用
Parameters
----------
moras : list[Mora]
モーラ系列
query : AudioQuery
音声合成用のクエリ
Returns
-------
moras : list[Mora]
話速スケールが適用されたモーラ系列
"""
"""モーラ系列へ音声合成用のクエリがもつ話速スケール（`speedScale`）を適用する"""
for mora in moras:
mora.vowel_length /= query.speedScale
if mora.consonant_length:
Expand Down Expand Up @@ -209,38 +185,14 @@ def calc_frame_per_mora(mora: Mora) -> ndarray:


def apply_pitch_scale(moras: list[Mora], query: AudioQuery) -> list[Mora]:
"""
音高スケール（`pitchScale`）の適用
Parameters
----------
moras : list[Mora]
モーラ系列
query : AudioQuery
音声合成用のクエリ
Returns
-------
moras : list[Mora]
音高スケールが適用されたモーラ系列
"""
"""モーラ系列へ音声合成用のクエリがもつ音高スケール（`pitchScale`）を適用する"""
for mora in moras:
mora.pitch *= 2**query.pitchScale
return moras


def apply_intonation_scale(moras: list[Mora], query: AudioQuery) -> list[Mora]:
"""
抑揚スケール（`intonationScale`）の適用
Parameters
----------
moras : list[Mora]
モーラ系列
query : AudioQuery
音声合成用のクエリ
Returns
-------
moras : list[Mora]
抑揚スケールが適用されたモーラ系列
"""
"""モーラ系列へ音声合成用のクエリがもつ抑揚スケール（`intonationScale`）を適用する"""
# 有声音素 (f0>0) の平均値に対する乖離度をスケール
voiced = list(filter(lambda mora: mora.pitch > 0, moras))
mean_f0 = numpy.mean(list(map(lambda mora: mora.pitch, voiced))).item()
Expand Down Expand Up @@ -274,19 +226,7 @@ def calc_frame_pitch(moras: list[Mora]) -> ndarray:


def apply_volume_scale(wave: numpy.ndarray, query: AudioQuery) -> numpy.ndarray:
"""
音量スケール（`volumeScale`）の適用
Parameters
----------
wave : numpy.ndarray
音声波形
query : AudioQuery
音声合成用のクエリ
Returns
-------
wave : numpy.ndarray
音量スケールが適用された音声波形
"""
"""音声波形へ音声合成用のクエリがもつ音量スケール（`volumeScale`）を適用する"""
wave *= query.volumeScale
return wave

Expand Down Expand Up @@ -317,43 +257,16 @@ def calc_frame_phoneme(phonemes: List[OjtPhoneme], frame_per_phoneme: numpy.ndar
def apply_output_sampling_rate(
wave: ndarray, sr_wave: int, query: AudioQuery
) -> ndarray:
"""
出力サンプリングレート（`outputSamplingRate`）の適用
Parameters
----------
wave : ndarray
音声波形
sr_wave : int
`wave`のサンプリングレート
query : AudioQuery
音声合成用のクエリ
Returns
-------
wave : ndarray
出力サンプリングレートが適用された音声波形
"""
"""音声波形へ音声合成用のクエリがもつ出力サンプリングレート（`outputSamplingRate`）を適用する"""
# サンプリングレート一致のときはスルー
if sr_wave == query.outputSamplingRate:
return wave

wave = resample(wave, sr_wave, query.outputSamplingRate)
return wave


def apply_output_stereo(wave: ndarray, query: AudioQuery) -> ndarray:
"""
ステレオ出力（`outputStereo`）の適用
Parameters
----------
wave : ndarray
音声波形
query : AudioQuery
音声合成用のクエリ
Returns
-------
wave : ndarray
ステレオ出力設定が適用された音声波形
"""
"""音声波形へ音声合成用のクエリがもつステレオ出力設定（`outputStereo`）を適用する"""
if query.outputStereo:
wave = numpy.array([wave, wave]).T
return wave
Expand All @@ -373,20 +286,20 @@ def query_to_decoder_feature(query: AudioQuery) -> tuple[ndarray, ndarray]:
f0 : ndarray
フレームごとの基本周波数、shape=(Frame,)
"""
flatten_moras = to_flatten_moras(query.accent_phrases)
moras = to_flatten_moras(query.accent_phrases)

flatten_moras = apply_prepost_silence(flatten_moras, query)
flatten_moras = apply_speed_scale(flatten_moras, query)
flatten_moras = apply_pitch_scale(flatten_moras, query)
flatten_moras = apply_intonation_scale(flatten_moras, query)
moras = apply_prepost_silence(moras, query)
moras = apply_speed_scale(moras, query)
moras = apply_pitch_scale(moras, query)
moras = apply_intonation_scale(moras, query)

phoneme_data_list = to_flatten_phonemes(flatten_moras)
phonemes = to_flatten_phonemes(moras)

frame_per_phoneme = calc_frame_per_phoneme(flatten_moras)
f0 = calc_frame_pitch(flatten_moras)
phoneme = calc_frame_phoneme(phoneme_data_list, frame_per_phoneme)
frame_per_phoneme = calc_frame_per_phoneme(moras)
f0 = calc_frame_pitch(moras)
phonemes = calc_frame_phoneme(phonemes, frame_per_phoneme)

return phoneme, f0
return phonemes, f0
Hiroshiba marked this conversation as resolved.
Show resolved Hide resolved


def raw_wave_to_output_wave(query: AudioQuery, wave: ndarray, sr_wave: int) -> ndarray:
Expand Down