Skip to content

Update TTS pronunciation dictionaries #15857

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

Open
quapham wants to merge 4 commits into
base: main
Choose a base branch
from
Open

Update TTS pronunciation dictionaries #15857

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
306f1fa
Update TTS pronunciation dictionaries
quapham Jun 30, 2026
8ebea33
Update IPA CMU dictionary version to nv26.07
quapham Jul 9, 2026
5a31913
Fix legacy MagpieTTS IPA dictionary path migration
quapham Jul 18, 2026
a97b186
Update test tolerances
blisc Jul 20, 2026
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
4 changes: 2 additions & 2 deletions .secrets.baseline
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1202,10 +1202,10 @@
"line_number": 9148
}
],
"scripts/tts_dataset_files/ipa_cmudict-0.7b_nv23.01.txt": [
"scripts/tts_dataset_files/ipa_cmudict-0.7b_nv26.07.txt": [
{
"type": "Artifactory Credentials",
"filename": "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv23.01.txt",
"filename": "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv26.07.txt",
"hashed_secret": "27b998c0976876189b861934085bba2964e49a11",
"is_verified": false,
"line_number": 4886
Expand Down
2 changes: 1 addition & 1 deletion examples/tts/conf/fastpitch_align_ipa.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,7 +28,7 @@ lowfreq: 0
highfreq: 8000
window: hann

phoneme_dict_path: "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv23.01.txt"
phoneme_dict_path: "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv26.07.txt"
heteronyms_path: "scripts/tts_dataset_files/heteronyms-052722"

model:
Expand Down
2 changes: 1 addition & 1 deletion examples/tts/conf/fastpitch_align_ipa_adapter.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,7 +29,7 @@ lowfreq: 0
highfreq: 8000
window: hann

phoneme_dict_path: "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv23.01.txt"
phoneme_dict_path: "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv26.07.txt"
heteronyms_path: "scripts/tts_dataset_files/heteronyms-052722"

reference_audio_type: "same-speaker" # options: ["same-speaker", "ground-truth"]
Expand Down
2 changes: 1 addition & 1 deletion examples/tts/conf/magpietts/easy_magpietts.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -109,7 +109,7 @@ model:
pronunciation_control_g2p:
en:
_target_: nemo.collections.tts.g2p.models.i18n_ipa.IpaG2p
phoneme_dict: "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv23.01.txt"
phoneme_dict: "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv26.07.txt"
heteronyms: "scripts/tts_dataset_files/heteronyms-052722"
phoneme_probability: 0.3
ignore_ambiguous_words: false
Expand Down
2 changes: 1 addition & 1 deletion examples/tts/conf/magpietts/easy_magpietts_lhotse.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -103,7 +103,7 @@ model:
pronunciation_control_g2p:
en:
_target_: nemo.collections.tts.g2p.models.i18n_ipa.IpaG2p
phoneme_dict: "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv23.01.txt"
phoneme_dict: "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv26.07.txt"
heteronyms: "scripts/tts_dataset_files/heteronyms-052722"
phoneme_probability: 0.3
ignore_ambiguous_words: false
Expand Down
2 changes: 1 addition & 1 deletion examples/tts/conf/magpietts/magpietts.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -65,7 +65,7 @@ model:
pad_with_space: false
g2p:
_target_: nemo.collections.tts.g2p.models.i18n_ipa.IpaG2p
phoneme_dict: "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv23.01.txt"
phoneme_dict: "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv26.07.txt"
heteronyms: "scripts/tts_dataset_files/heteronyms-052722"
phoneme_probability: 0.8
ignore_ambiguous_words: false
Expand Down
2 changes: 1 addition & 1 deletion examples/tts/conf/magpietts/magpietts_lhotse.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -54,7 +54,7 @@ model:
pad_with_space: false
g2p:
_target_: nemo.collections.tts.g2p.models.i18n_ipa.IpaG2p
phoneme_dict: "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv23.01.txt"
phoneme_dict: "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv26.07.txt"
heteronyms: "scripts/tts_dataset_files/heteronyms-052722"
phoneme_probability: 0.8
ignore_ambiguous_words: false
Expand Down
2 changes: 1 addition & 1 deletion examples/tts/conf/magpietts/magpietts_lhotse_moe.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -57,7 +57,7 @@ model:
pad_with_space: false
g2p:
_target_: nemo.collections.tts.g2p.models.i18n_ipa.IpaG2p
phoneme_dict: "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv23.01.txt"
phoneme_dict: "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv26.07.txt"
heteronyms: "scripts/tts_dataset_files/heteronyms-052722"
phoneme_probability: 0.8
ignore_ambiguous_words: false
Expand Down
4 changes: 2 additions & 2 deletions examples/tts/conf/magpietts/magpietts_po_inference.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@ weighted_sampling_steps_per_epoch: null
# See DatasetMeta in https://github.com/NVIDIA-NeMo/NeMo/blob/main/nemo/collections/tts/data/text_to_speech_dataset.py
test_ds_meta: ???

phoneme_dict_path: "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv23.01.txt"
phoneme_dict_path: "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv26.07.txt"
heteronyms_path: "scripts/tts_dataset_files/heteronyms-052722"
model:
# Inference hyperparameters
Expand Down Expand Up @@ -73,7 +73,7 @@ model:
pad_with_space: false
g2p:
_target_: nemo.collections.tts.g2p.models.i18n_ipa.IpaG2p
phoneme_dict: "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv23.01.txt"
phoneme_dict: "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv26.07.txt"
heteronyms: "scripts/tts_dataset_files/heteronyms-052722"
phoneme_probability: 0.8
ignore_ambiguous_words: false
Expand Down
2 changes: 1 addition & 1 deletion nemo/collections/tts/modules/magpietts_inference/utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -228,7 +228,7 @@ def update_config_for_inference(

# Update text tokenizer paths for backward compatibility
if hasattr(model_cfg, 'text_tokenizer'):
model_cfg.text_tokenizer.g2p.phoneme_dict = "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv23.01.txt"
model_cfg.text_tokenizer.g2p.phoneme_dict = "scripts/tts_dataset_files/ipa_cmudict-0.7b_nv26.07.txt"
model_cfg.text_tokenizer.g2p.heteronyms = "scripts/tts_dataset_files/heteronyms-052722"
model_cfg.text_tokenizer.g2p.phoneme_probability = 1.0

Expand Down
Loading
Loading