diff --git a/README.md b/README.md index aca4521..ccf60bf 100644 --- a/README.md +++ b/README.md @@ -84,7 +84,7 @@ text_prompt = """ I have a silky smooth voice, and today I will tell you about the exercise regimen of the common sloth. """ -audio_array = generate_audio(text_prompt, history_prompt="speech_0") +audio_array = generate_audio(text_prompt, history_prompt="en_speaker_1") ``` [sloth.webm](https://user-images.githubusercontent.com/5068315/230684883-a344c619-a560-4ff5-8b99-b4463a34487b.webm) diff --git a/bark/assets/prompts/speech_0.npz b/bark/assets/prompts/announcer.npz similarity index 100% rename from bark/assets/prompts/speech_0.npz rename to bark/assets/prompts/announcer.npz diff --git a/bark/assets/prompts/de_speaker_0.npz b/bark/assets/prompts/de_speaker_0.npz new file mode 100644 index 0000000..04d6148 Binary files /dev/null and b/bark/assets/prompts/de_speaker_0.npz differ diff --git a/bark/assets/prompts/de_speaker_1.npz b/bark/assets/prompts/de_speaker_1.npz new file mode 100644 index 0000000..f733052 Binary files /dev/null and b/bark/assets/prompts/de_speaker_1.npz differ diff --git a/bark/assets/prompts/de_speaker_2.npz b/bark/assets/prompts/de_speaker_2.npz new file mode 100644 index 0000000..4b9684d Binary files /dev/null and b/bark/assets/prompts/de_speaker_2.npz differ diff --git a/bark/assets/prompts/de_speaker_3.npz b/bark/assets/prompts/de_speaker_3.npz new file mode 100644 index 0000000..4d70146 Binary files /dev/null and b/bark/assets/prompts/de_speaker_3.npz differ diff --git a/bark/assets/prompts/de_speaker_4.npz b/bark/assets/prompts/de_speaker_4.npz new file mode 100644 index 0000000..9b8c2db Binary files /dev/null and b/bark/assets/prompts/de_speaker_4.npz differ diff --git a/bark/assets/prompts/de_speaker_5.npz b/bark/assets/prompts/de_speaker_5.npz new file mode 100644 index 0000000..0491721 Binary files /dev/null and b/bark/assets/prompts/de_speaker_5.npz differ diff --git a/bark/assets/prompts/de_speaker_6.npz b/bark/assets/prompts/de_speaker_6.npz new file mode 100644 index 0000000..dcfc2f3 Binary files /dev/null and b/bark/assets/prompts/de_speaker_6.npz differ diff --git a/bark/assets/prompts/de_speaker_7.npz b/bark/assets/prompts/de_speaker_7.npz new file mode 100644 index 0000000..c1af725 Binary files /dev/null and b/bark/assets/prompts/de_speaker_7.npz differ diff --git a/bark/assets/prompts/de_speaker_8.npz b/bark/assets/prompts/de_speaker_8.npz new file mode 100644 index 0000000..0c973f5 Binary files /dev/null and b/bark/assets/prompts/de_speaker_8.npz differ diff --git a/bark/assets/prompts/de_speaker_9.npz b/bark/assets/prompts/de_speaker_9.npz new file mode 100644 index 0000000..733ab92 Binary files /dev/null and b/bark/assets/prompts/de_speaker_9.npz differ diff --git a/bark/assets/prompts/en_speaker_0.npz b/bark/assets/prompts/en_speaker_0.npz new file mode 100644 index 0000000..dffc3ce Binary files /dev/null and b/bark/assets/prompts/en_speaker_0.npz differ diff --git a/bark/assets/prompts/en_speaker_1.npz b/bark/assets/prompts/en_speaker_1.npz new file mode 100644 index 0000000..903460f Binary files /dev/null and b/bark/assets/prompts/en_speaker_1.npz differ diff --git a/bark/assets/prompts/en_speaker_2.npz b/bark/assets/prompts/en_speaker_2.npz new file mode 100644 index 0000000..8ba83d9 Binary files /dev/null and b/bark/assets/prompts/en_speaker_2.npz differ diff --git a/bark/assets/prompts/en_speaker_3.npz b/bark/assets/prompts/en_speaker_3.npz new file mode 100644 index 0000000..8f2c96e Binary files /dev/null and b/bark/assets/prompts/en_speaker_3.npz differ diff --git a/bark/assets/prompts/en_speaker_4.npz b/bark/assets/prompts/en_speaker_4.npz new file mode 100644 index 0000000..eaf26cd Binary files /dev/null and b/bark/assets/prompts/en_speaker_4.npz differ diff --git a/bark/assets/prompts/en_speaker_5.npz b/bark/assets/prompts/en_speaker_5.npz new file mode 100644 index 0000000..55a8e82 Binary files /dev/null and b/bark/assets/prompts/en_speaker_5.npz differ diff --git a/bark/assets/prompts/en_speaker_6.npz b/bark/assets/prompts/en_speaker_6.npz new file mode 100644 index 0000000..ddbdd11 Binary files /dev/null and b/bark/assets/prompts/en_speaker_6.npz differ diff --git a/bark/assets/prompts/en_speaker_7.npz b/bark/assets/prompts/en_speaker_7.npz new file mode 100644 index 0000000..456453c Binary files /dev/null and b/bark/assets/prompts/en_speaker_7.npz differ diff --git a/bark/assets/prompts/en_speaker_8.npz b/bark/assets/prompts/en_speaker_8.npz new file mode 100644 index 0000000..ec9b77a Binary files /dev/null and b/bark/assets/prompts/en_speaker_8.npz differ diff --git a/bark/assets/prompts/en_speaker_9.npz b/bark/assets/prompts/en_speaker_9.npz new file mode 100644 index 0000000..0ff7d58 Binary files /dev/null and b/bark/assets/prompts/en_speaker_9.npz differ diff --git a/bark/assets/prompts/es_speaker_0.npz b/bark/assets/prompts/es_speaker_0.npz new file mode 100644 index 0000000..218920e Binary files /dev/null and b/bark/assets/prompts/es_speaker_0.npz differ diff --git a/bark/assets/prompts/es_speaker_1.npz b/bark/assets/prompts/es_speaker_1.npz new file mode 100644 index 0000000..f5ac7d7 Binary files /dev/null and b/bark/assets/prompts/es_speaker_1.npz differ diff --git a/bark/assets/prompts/es_speaker_2.npz b/bark/assets/prompts/es_speaker_2.npz new file mode 100644 index 0000000..9915e34 Binary files /dev/null and b/bark/assets/prompts/es_speaker_2.npz differ diff --git a/bark/assets/prompts/es_speaker_3.npz b/bark/assets/prompts/es_speaker_3.npz new file mode 100644 index 0000000..e20757b Binary files /dev/null and b/bark/assets/prompts/es_speaker_3.npz differ diff --git a/bark/assets/prompts/es_speaker_4.npz b/bark/assets/prompts/es_speaker_4.npz new file mode 100644 index 0000000..3bcbe4f Binary files /dev/null and b/bark/assets/prompts/es_speaker_4.npz differ diff --git a/bark/assets/prompts/es_speaker_5.npz b/bark/assets/prompts/es_speaker_5.npz new file mode 100644 index 0000000..654a8b8 Binary files /dev/null and b/bark/assets/prompts/es_speaker_5.npz differ diff --git a/bark/assets/prompts/es_speaker_6.npz b/bark/assets/prompts/es_speaker_6.npz new file mode 100644 index 0000000..034f9a8 Binary files /dev/null and b/bark/assets/prompts/es_speaker_6.npz differ diff --git a/bark/assets/prompts/es_speaker_7.npz b/bark/assets/prompts/es_speaker_7.npz new file mode 100644 index 0000000..ac76151 Binary files /dev/null and b/bark/assets/prompts/es_speaker_7.npz differ diff --git a/bark/assets/prompts/es_speaker_8.npz b/bark/assets/prompts/es_speaker_8.npz new file mode 100644 index 0000000..40715bb Binary files /dev/null and b/bark/assets/prompts/es_speaker_8.npz differ diff --git a/bark/assets/prompts/es_speaker_9.npz b/bark/assets/prompts/es_speaker_9.npz new file mode 100644 index 0000000..28875e0 Binary files /dev/null and b/bark/assets/prompts/es_speaker_9.npz differ diff --git a/bark/assets/prompts/fr_speaker_0.npz b/bark/assets/prompts/fr_speaker_0.npz new file mode 100644 index 0000000..1b92a37 Binary files /dev/null and b/bark/assets/prompts/fr_speaker_0.npz differ diff --git a/bark/assets/prompts/fr_speaker_1.npz b/bark/assets/prompts/fr_speaker_1.npz new file mode 100644 index 0000000..85ec6f4 Binary files /dev/null and b/bark/assets/prompts/fr_speaker_1.npz differ diff --git a/bark/assets/prompts/fr_speaker_2.npz b/bark/assets/prompts/fr_speaker_2.npz new file mode 100644 index 0000000..7510c4e Binary files /dev/null and b/bark/assets/prompts/fr_speaker_2.npz differ diff --git a/bark/assets/prompts/fr_speaker_3.npz b/bark/assets/prompts/fr_speaker_3.npz new file mode 100644 index 0000000..e905409 Binary files /dev/null and b/bark/assets/prompts/fr_speaker_3.npz differ diff --git a/bark/assets/prompts/fr_speaker_4.npz b/bark/assets/prompts/fr_speaker_4.npz new file mode 100644 index 0000000..ed5717a Binary files /dev/null and b/bark/assets/prompts/fr_speaker_4.npz differ diff --git a/bark/assets/prompts/fr_speaker_5.npz b/bark/assets/prompts/fr_speaker_5.npz new file mode 100644 index 0000000..8c3e56c Binary files /dev/null and b/bark/assets/prompts/fr_speaker_5.npz differ diff --git a/bark/assets/prompts/fr_speaker_6.npz b/bark/assets/prompts/fr_speaker_6.npz new file mode 100644 index 0000000..fabed64 Binary files /dev/null and b/bark/assets/prompts/fr_speaker_6.npz differ diff --git a/bark/assets/prompts/fr_speaker_7.npz b/bark/assets/prompts/fr_speaker_7.npz new file mode 100644 index 0000000..e319d41 Binary files /dev/null and b/bark/assets/prompts/fr_speaker_7.npz differ diff --git a/bark/assets/prompts/fr_speaker_8.npz b/bark/assets/prompts/fr_speaker_8.npz new file mode 100644 index 0000000..e8d2b9e Binary files /dev/null and b/bark/assets/prompts/fr_speaker_8.npz differ diff --git a/bark/assets/prompts/fr_speaker_9.npz b/bark/assets/prompts/fr_speaker_9.npz new file mode 100644 index 0000000..fd552b2 Binary files /dev/null and b/bark/assets/prompts/fr_speaker_9.npz differ diff --git a/bark/assets/prompts/hi_speaker_0.npz b/bark/assets/prompts/hi_speaker_0.npz new file mode 100644 index 0000000..6fa2da1 Binary files /dev/null and b/bark/assets/prompts/hi_speaker_0.npz differ diff --git a/bark/assets/prompts/hi_speaker_1.npz b/bark/assets/prompts/hi_speaker_1.npz new file mode 100644 index 0000000..30b5a94 Binary files /dev/null and b/bark/assets/prompts/hi_speaker_1.npz differ diff --git a/bark/assets/prompts/hi_speaker_2.npz b/bark/assets/prompts/hi_speaker_2.npz new file mode 100644 index 0000000..abf79cb Binary files /dev/null and b/bark/assets/prompts/hi_speaker_2.npz differ diff --git a/bark/assets/prompts/hi_speaker_3.npz b/bark/assets/prompts/hi_speaker_3.npz new file mode 100644 index 0000000..3ab57ab Binary files /dev/null and b/bark/assets/prompts/hi_speaker_3.npz differ diff --git a/bark/assets/prompts/hi_speaker_4.npz b/bark/assets/prompts/hi_speaker_4.npz new file mode 100644 index 0000000..115a523 Binary files /dev/null and b/bark/assets/prompts/hi_speaker_4.npz differ diff --git a/bark/assets/prompts/hi_speaker_5.npz b/bark/assets/prompts/hi_speaker_5.npz new file mode 100644 index 0000000..b8fc580 Binary files /dev/null and b/bark/assets/prompts/hi_speaker_5.npz differ diff --git a/bark/assets/prompts/hi_speaker_6.npz b/bark/assets/prompts/hi_speaker_6.npz new file mode 100644 index 0000000..ac03556 Binary files /dev/null and b/bark/assets/prompts/hi_speaker_6.npz differ diff --git a/bark/assets/prompts/hi_speaker_7.npz b/bark/assets/prompts/hi_speaker_7.npz new file mode 100644 index 0000000..fae7302 Binary files /dev/null and b/bark/assets/prompts/hi_speaker_7.npz differ diff --git a/bark/assets/prompts/hi_speaker_8.npz b/bark/assets/prompts/hi_speaker_8.npz new file mode 100644 index 0000000..6d6f512 Binary files /dev/null and b/bark/assets/prompts/hi_speaker_8.npz differ diff --git a/bark/assets/prompts/hi_speaker_9.npz b/bark/assets/prompts/hi_speaker_9.npz new file mode 100644 index 0000000..ccc7905 Binary files /dev/null and b/bark/assets/prompts/hi_speaker_9.npz differ diff --git a/bark/assets/prompts/it_speaker_0.npz b/bark/assets/prompts/it_speaker_0.npz new file mode 100644 index 0000000..512ff36 Binary files /dev/null and b/bark/assets/prompts/it_speaker_0.npz differ diff --git a/bark/assets/prompts/it_speaker_1.npz b/bark/assets/prompts/it_speaker_1.npz new file mode 100644 index 0000000..256c357 Binary files /dev/null and b/bark/assets/prompts/it_speaker_1.npz differ diff --git a/bark/assets/prompts/it_speaker_2.npz b/bark/assets/prompts/it_speaker_2.npz new file mode 100644 index 0000000..c84df86 Binary files /dev/null and b/bark/assets/prompts/it_speaker_2.npz differ diff --git a/bark/assets/prompts/it_speaker_3.npz b/bark/assets/prompts/it_speaker_3.npz new file mode 100644 index 0000000..168840a Binary files /dev/null and b/bark/assets/prompts/it_speaker_3.npz differ diff --git a/bark/assets/prompts/it_speaker_4.npz b/bark/assets/prompts/it_speaker_4.npz new file mode 100644 index 0000000..e54bae9 Binary files /dev/null and b/bark/assets/prompts/it_speaker_4.npz differ diff --git a/bark/assets/prompts/it_speaker_5.npz b/bark/assets/prompts/it_speaker_5.npz new file mode 100644 index 0000000..e99ee6f Binary files /dev/null and b/bark/assets/prompts/it_speaker_5.npz differ diff --git a/bark/assets/prompts/it_speaker_6.npz b/bark/assets/prompts/it_speaker_6.npz new file mode 100644 index 0000000..cb11adb Binary files /dev/null and b/bark/assets/prompts/it_speaker_6.npz differ diff --git a/bark/assets/prompts/it_speaker_7.npz b/bark/assets/prompts/it_speaker_7.npz new file mode 100644 index 0000000..cbcd4ae Binary files /dev/null and b/bark/assets/prompts/it_speaker_7.npz differ diff --git a/bark/assets/prompts/it_speaker_8.npz b/bark/assets/prompts/it_speaker_8.npz new file mode 100644 index 0000000..34b224a Binary files /dev/null and b/bark/assets/prompts/it_speaker_8.npz differ diff --git a/bark/assets/prompts/it_speaker_9.npz b/bark/assets/prompts/it_speaker_9.npz new file mode 100644 index 0000000..3536913 Binary files /dev/null and b/bark/assets/prompts/it_speaker_9.npz differ diff --git a/bark/assets/prompts/ja_speaker_0.npz b/bark/assets/prompts/ja_speaker_0.npz new file mode 100644 index 0000000..ee4479a Binary files /dev/null and b/bark/assets/prompts/ja_speaker_0.npz differ diff --git a/bark/assets/prompts/ja_speaker_1.npz b/bark/assets/prompts/ja_speaker_1.npz new file mode 100644 index 0000000..a134f39 Binary files /dev/null and b/bark/assets/prompts/ja_speaker_1.npz differ diff --git a/bark/assets/prompts/ja_speaker_2.npz b/bark/assets/prompts/ja_speaker_2.npz new file mode 100644 index 0000000..adde4ec Binary files /dev/null and b/bark/assets/prompts/ja_speaker_2.npz differ diff --git a/bark/assets/prompts/ja_speaker_3.npz b/bark/assets/prompts/ja_speaker_3.npz new file mode 100644 index 0000000..9c08f76 Binary files /dev/null and b/bark/assets/prompts/ja_speaker_3.npz differ diff --git a/bark/assets/prompts/ja_speaker_4.npz b/bark/assets/prompts/ja_speaker_4.npz new file mode 100644 index 0000000..981aa05 Binary files /dev/null and b/bark/assets/prompts/ja_speaker_4.npz differ diff --git a/bark/assets/prompts/ja_speaker_5.npz b/bark/assets/prompts/ja_speaker_5.npz new file mode 100644 index 0000000..38578be Binary files /dev/null and b/bark/assets/prompts/ja_speaker_5.npz differ diff --git a/bark/assets/prompts/ja_speaker_6.npz b/bark/assets/prompts/ja_speaker_6.npz new file mode 100644 index 0000000..2f016e7 Binary files /dev/null and b/bark/assets/prompts/ja_speaker_6.npz differ diff --git a/bark/assets/prompts/ja_speaker_7.npz b/bark/assets/prompts/ja_speaker_7.npz new file mode 100644 index 0000000..5bc5c84 Binary files /dev/null and b/bark/assets/prompts/ja_speaker_7.npz differ diff --git a/bark/assets/prompts/ja_speaker_8.npz b/bark/assets/prompts/ja_speaker_8.npz new file mode 100644 index 0000000..206953b Binary files /dev/null and b/bark/assets/prompts/ja_speaker_8.npz differ diff --git a/bark/assets/prompts/ja_speaker_9.npz b/bark/assets/prompts/ja_speaker_9.npz new file mode 100644 index 0000000..7216e42 Binary files /dev/null and b/bark/assets/prompts/ja_speaker_9.npz differ diff --git a/bark/assets/prompts/ko_speaker_0.npz b/bark/assets/prompts/ko_speaker_0.npz new file mode 100644 index 0000000..fd40028 Binary files /dev/null and b/bark/assets/prompts/ko_speaker_0.npz differ diff --git a/bark/assets/prompts/ko_speaker_1.npz b/bark/assets/prompts/ko_speaker_1.npz new file mode 100644 index 0000000..0749da1 Binary files /dev/null and b/bark/assets/prompts/ko_speaker_1.npz differ diff --git a/bark/assets/prompts/ko_speaker_2.npz b/bark/assets/prompts/ko_speaker_2.npz new file mode 100644 index 0000000..19a35c8 Binary files /dev/null and b/bark/assets/prompts/ko_speaker_2.npz differ diff --git a/bark/assets/prompts/ko_speaker_3.npz b/bark/assets/prompts/ko_speaker_3.npz new file mode 100644 index 0000000..862986e Binary files /dev/null and b/bark/assets/prompts/ko_speaker_3.npz differ diff --git a/bark/assets/prompts/ko_speaker_4.npz b/bark/assets/prompts/ko_speaker_4.npz new file mode 100644 index 0000000..a1b1b84 Binary files /dev/null and b/bark/assets/prompts/ko_speaker_4.npz differ diff --git a/bark/assets/prompts/ko_speaker_5.npz b/bark/assets/prompts/ko_speaker_5.npz new file mode 100644 index 0000000..59b110a Binary files /dev/null and b/bark/assets/prompts/ko_speaker_5.npz differ diff --git a/bark/assets/prompts/ko_speaker_6.npz b/bark/assets/prompts/ko_speaker_6.npz new file mode 100644 index 0000000..053ccc3 Binary files /dev/null and b/bark/assets/prompts/ko_speaker_6.npz differ diff --git a/bark/assets/prompts/ko_speaker_7.npz b/bark/assets/prompts/ko_speaker_7.npz new file mode 100644 index 0000000..49f268e Binary files /dev/null and b/bark/assets/prompts/ko_speaker_7.npz differ diff --git a/bark/assets/prompts/ko_speaker_8.npz b/bark/assets/prompts/ko_speaker_8.npz new file mode 100644 index 0000000..0400ee4 Binary files /dev/null and b/bark/assets/prompts/ko_speaker_8.npz differ diff --git a/bark/assets/prompts/ko_speaker_9.npz b/bark/assets/prompts/ko_speaker_9.npz new file mode 100644 index 0000000..270b090 Binary files /dev/null and b/bark/assets/prompts/ko_speaker_9.npz differ diff --git a/bark/assets/prompts/music_0.npz b/bark/assets/prompts/music_0.npz deleted file mode 100644 index 4701038..0000000 Binary files a/bark/assets/prompts/music_0.npz and /dev/null differ diff --git a/bark/assets/prompts/music_1.npz b/bark/assets/prompts/music_1.npz deleted file mode 100644 index b95f992..0000000 Binary files a/bark/assets/prompts/music_1.npz and /dev/null differ diff --git a/bark/assets/prompts/music_2.npz b/bark/assets/prompts/music_2.npz deleted file mode 100644 index 94c2577..0000000 Binary files a/bark/assets/prompts/music_2.npz and /dev/null differ diff --git a/bark/assets/prompts/music_3.npz b/bark/assets/prompts/music_3.npz deleted file mode 100644 index c79219d..0000000 Binary files a/bark/assets/prompts/music_3.npz and /dev/null differ diff --git a/bark/assets/prompts/music_4.npz b/bark/assets/prompts/music_4.npz deleted file mode 100644 index 5053865..0000000 Binary files a/bark/assets/prompts/music_4.npz and /dev/null differ diff --git a/bark/assets/prompts/music_5.npz b/bark/assets/prompts/music_5.npz deleted file mode 100644 index 5c96632..0000000 Binary files a/bark/assets/prompts/music_5.npz and /dev/null differ diff --git a/bark/assets/prompts/pl_speaker_0.npz b/bark/assets/prompts/pl_speaker_0.npz new file mode 100644 index 0000000..b06af76 Binary files /dev/null and b/bark/assets/prompts/pl_speaker_0.npz differ diff --git a/bark/assets/prompts/pl_speaker_1.npz b/bark/assets/prompts/pl_speaker_1.npz new file mode 100644 index 0000000..b93ac1a Binary files /dev/null and b/bark/assets/prompts/pl_speaker_1.npz differ diff --git a/bark/assets/prompts/pl_speaker_2.npz b/bark/assets/prompts/pl_speaker_2.npz new file mode 100644 index 0000000..d52643e Binary files /dev/null and b/bark/assets/prompts/pl_speaker_2.npz differ diff --git a/bark/assets/prompts/pl_speaker_3.npz b/bark/assets/prompts/pl_speaker_3.npz new file mode 100644 index 0000000..31969bb Binary files /dev/null and b/bark/assets/prompts/pl_speaker_3.npz differ diff --git a/bark/assets/prompts/pl_speaker_4.npz b/bark/assets/prompts/pl_speaker_4.npz new file mode 100644 index 0000000..56b320b Binary files /dev/null and b/bark/assets/prompts/pl_speaker_4.npz differ diff --git a/bark/assets/prompts/pl_speaker_5.npz b/bark/assets/prompts/pl_speaker_5.npz new file mode 100644 index 0000000..7097600 Binary files /dev/null and b/bark/assets/prompts/pl_speaker_5.npz differ diff --git a/bark/assets/prompts/pl_speaker_6.npz b/bark/assets/prompts/pl_speaker_6.npz new file mode 100644 index 0000000..277ccc4 Binary files /dev/null and b/bark/assets/prompts/pl_speaker_6.npz differ diff --git a/bark/assets/prompts/pl_speaker_7.npz b/bark/assets/prompts/pl_speaker_7.npz new file mode 100644 index 0000000..e6a1626 Binary files /dev/null and b/bark/assets/prompts/pl_speaker_7.npz differ diff --git a/bark/assets/prompts/pl_speaker_8.npz b/bark/assets/prompts/pl_speaker_8.npz new file mode 100644 index 0000000..6f733de Binary files /dev/null and b/bark/assets/prompts/pl_speaker_8.npz differ diff --git a/bark/assets/prompts/pl_speaker_9.npz b/bark/assets/prompts/pl_speaker_9.npz new file mode 100644 index 0000000..b5913a6 Binary files /dev/null and b/bark/assets/prompts/pl_speaker_9.npz differ diff --git a/bark/assets/prompts/pt_speaker_0.npz b/bark/assets/prompts/pt_speaker_0.npz new file mode 100644 index 0000000..7d2ef2f Binary files /dev/null and b/bark/assets/prompts/pt_speaker_0.npz differ diff --git a/bark/assets/prompts/pt_speaker_1.npz b/bark/assets/prompts/pt_speaker_1.npz new file mode 100644 index 0000000..ea093f2 Binary files /dev/null and b/bark/assets/prompts/pt_speaker_1.npz differ diff --git a/bark/assets/prompts/pt_speaker_2.npz b/bark/assets/prompts/pt_speaker_2.npz new file mode 100644 index 0000000..a5fdb00 Binary files /dev/null and b/bark/assets/prompts/pt_speaker_2.npz differ diff --git a/bark/assets/prompts/pt_speaker_3.npz b/bark/assets/prompts/pt_speaker_3.npz new file mode 100644 index 0000000..6b2742d Binary files /dev/null and b/bark/assets/prompts/pt_speaker_3.npz differ diff --git a/bark/assets/prompts/pt_speaker_4.npz b/bark/assets/prompts/pt_speaker_4.npz new file mode 100644 index 0000000..7836e83 Binary files /dev/null and b/bark/assets/prompts/pt_speaker_4.npz differ diff --git a/bark/assets/prompts/pt_speaker_5.npz b/bark/assets/prompts/pt_speaker_5.npz new file mode 100644 index 0000000..4f99747 Binary files /dev/null and b/bark/assets/prompts/pt_speaker_5.npz differ diff --git a/bark/assets/prompts/pt_speaker_6.npz b/bark/assets/prompts/pt_speaker_6.npz new file mode 100644 index 0000000..0e7fb7a Binary files /dev/null and b/bark/assets/prompts/pt_speaker_6.npz differ diff --git a/bark/assets/prompts/pt_speaker_7.npz b/bark/assets/prompts/pt_speaker_7.npz new file mode 100644 index 0000000..1205123 Binary files /dev/null and b/bark/assets/prompts/pt_speaker_7.npz differ diff --git a/bark/assets/prompts/pt_speaker_8.npz b/bark/assets/prompts/pt_speaker_8.npz new file mode 100644 index 0000000..e3f46a5 Binary files /dev/null and b/bark/assets/prompts/pt_speaker_8.npz differ diff --git a/bark/assets/prompts/pt_speaker_9.npz b/bark/assets/prompts/pt_speaker_9.npz new file mode 100644 index 0000000..512a231 Binary files /dev/null and b/bark/assets/prompts/pt_speaker_9.npz differ diff --git a/bark/assets/prompts/ru_speaker_0.npz b/bark/assets/prompts/ru_speaker_0.npz new file mode 100644 index 0000000..00d6c51 Binary files /dev/null and b/bark/assets/prompts/ru_speaker_0.npz differ diff --git a/bark/assets/prompts/ru_speaker_1.npz b/bark/assets/prompts/ru_speaker_1.npz new file mode 100644 index 0000000..420a25d Binary files /dev/null and b/bark/assets/prompts/ru_speaker_1.npz differ diff --git a/bark/assets/prompts/ru_speaker_2.npz b/bark/assets/prompts/ru_speaker_2.npz new file mode 100644 index 0000000..efdb07c Binary files /dev/null and b/bark/assets/prompts/ru_speaker_2.npz differ diff --git a/bark/assets/prompts/ru_speaker_3.npz b/bark/assets/prompts/ru_speaker_3.npz new file mode 100644 index 0000000..cb0275f Binary files /dev/null and b/bark/assets/prompts/ru_speaker_3.npz differ diff --git a/bark/assets/prompts/ru_speaker_4.npz b/bark/assets/prompts/ru_speaker_4.npz new file mode 100644 index 0000000..0def7aa Binary files /dev/null and b/bark/assets/prompts/ru_speaker_4.npz differ diff --git a/bark/assets/prompts/ru_speaker_5.npz b/bark/assets/prompts/ru_speaker_5.npz new file mode 100644 index 0000000..52dae0e Binary files /dev/null and b/bark/assets/prompts/ru_speaker_5.npz differ diff --git a/bark/assets/prompts/ru_speaker_6.npz b/bark/assets/prompts/ru_speaker_6.npz new file mode 100644 index 0000000..86cad7a Binary files /dev/null and b/bark/assets/prompts/ru_speaker_6.npz differ diff --git a/bark/assets/prompts/ru_speaker_7.npz b/bark/assets/prompts/ru_speaker_7.npz new file mode 100644 index 0000000..c487a60 Binary files /dev/null and b/bark/assets/prompts/ru_speaker_7.npz differ diff --git a/bark/assets/prompts/ru_speaker_8.npz b/bark/assets/prompts/ru_speaker_8.npz new file mode 100644 index 0000000..311dc08 Binary files /dev/null and b/bark/assets/prompts/ru_speaker_8.npz differ diff --git a/bark/assets/prompts/ru_speaker_9.npz b/bark/assets/prompts/ru_speaker_9.npz new file mode 100644 index 0000000..df96368 Binary files /dev/null and b/bark/assets/prompts/ru_speaker_9.npz differ diff --git a/bark/assets/prompts/speaker_0.npz b/bark/assets/prompts/speaker_0.npz new file mode 100644 index 0000000..c6181ab Binary files /dev/null and b/bark/assets/prompts/speaker_0.npz differ diff --git a/bark/assets/prompts/speaker_1.npz b/bark/assets/prompts/speaker_1.npz new file mode 100644 index 0000000..8ce81fc Binary files /dev/null and b/bark/assets/prompts/speaker_1.npz differ diff --git a/bark/assets/prompts/speaker_2.npz b/bark/assets/prompts/speaker_2.npz new file mode 100644 index 0000000..289fbb9 Binary files /dev/null and b/bark/assets/prompts/speaker_2.npz differ diff --git a/bark/assets/prompts/speaker_3.npz b/bark/assets/prompts/speaker_3.npz new file mode 100644 index 0000000..02c352d Binary files /dev/null and b/bark/assets/prompts/speaker_3.npz differ diff --git a/bark/assets/prompts/speaker_4.npz b/bark/assets/prompts/speaker_4.npz new file mode 100644 index 0000000..b696cf3 Binary files /dev/null and b/bark/assets/prompts/speaker_4.npz differ diff --git a/bark/assets/prompts/speaker_5.npz b/bark/assets/prompts/speaker_5.npz new file mode 100644 index 0000000..5770a70 Binary files /dev/null and b/bark/assets/prompts/speaker_5.npz differ diff --git a/bark/assets/prompts/speaker_6.npz b/bark/assets/prompts/speaker_6.npz new file mode 100644 index 0000000..1fa0e65 Binary files /dev/null and b/bark/assets/prompts/speaker_6.npz differ diff --git a/bark/assets/prompts/speaker_7.npz b/bark/assets/prompts/speaker_7.npz new file mode 100644 index 0000000..50336bd Binary files /dev/null and b/bark/assets/prompts/speaker_7.npz differ diff --git a/bark/assets/prompts/speaker_8.npz b/bark/assets/prompts/speaker_8.npz new file mode 100644 index 0000000..44d2218 Binary files /dev/null and b/bark/assets/prompts/speaker_8.npz differ diff --git a/bark/assets/prompts/speaker_9.npz b/bark/assets/prompts/speaker_9.npz new file mode 100644 index 0000000..9d929ba Binary files /dev/null and b/bark/assets/prompts/speaker_9.npz differ diff --git a/bark/assets/prompts/speech_1.npz b/bark/assets/prompts/speech_1.npz deleted file mode 100644 index c44d276..0000000 Binary files a/bark/assets/prompts/speech_1.npz and /dev/null differ diff --git a/bark/assets/prompts/speech_2.npz b/bark/assets/prompts/speech_2.npz deleted file mode 100644 index 56843f3..0000000 Binary files a/bark/assets/prompts/speech_2.npz and /dev/null differ diff --git a/bark/assets/prompts/speech_3.npz b/bark/assets/prompts/speech_3.npz deleted file mode 100644 index d821f5d..0000000 Binary files a/bark/assets/prompts/speech_3.npz and /dev/null differ diff --git a/bark/assets/prompts/speech_4.npz b/bark/assets/prompts/speech_4.npz deleted file mode 100644 index 0e82fe0..0000000 Binary files a/bark/assets/prompts/speech_4.npz and /dev/null differ diff --git a/bark/assets/prompts/speech_5.npz b/bark/assets/prompts/speech_5.npz deleted file mode 100644 index 3f90b5d..0000000 Binary files a/bark/assets/prompts/speech_5.npz and /dev/null differ diff --git a/bark/assets/prompts/speech_6.npz b/bark/assets/prompts/speech_6.npz deleted file mode 100644 index 88c1787..0000000 Binary files a/bark/assets/prompts/speech_6.npz and /dev/null differ diff --git a/bark/assets/prompts/speech_7.npz b/bark/assets/prompts/speech_7.npz deleted file mode 100644 index fd32c3b..0000000 Binary files a/bark/assets/prompts/speech_7.npz and /dev/null differ diff --git a/bark/assets/prompts/tr_speaker_0.npz b/bark/assets/prompts/tr_speaker_0.npz new file mode 100644 index 0000000..e22c49c Binary files /dev/null and b/bark/assets/prompts/tr_speaker_0.npz differ diff --git a/bark/assets/prompts/tr_speaker_1.npz b/bark/assets/prompts/tr_speaker_1.npz new file mode 100644 index 0000000..47e4f75 Binary files /dev/null and b/bark/assets/prompts/tr_speaker_1.npz differ diff --git a/bark/assets/prompts/tr_speaker_2.npz b/bark/assets/prompts/tr_speaker_2.npz new file mode 100644 index 0000000..f6a02f9 Binary files /dev/null and b/bark/assets/prompts/tr_speaker_2.npz differ diff --git a/bark/assets/prompts/tr_speaker_3.npz b/bark/assets/prompts/tr_speaker_3.npz new file mode 100644 index 0000000..e5a22fe Binary files /dev/null and b/bark/assets/prompts/tr_speaker_3.npz differ diff --git a/bark/assets/prompts/tr_speaker_4.npz b/bark/assets/prompts/tr_speaker_4.npz new file mode 100644 index 0000000..f414e7e Binary files /dev/null and b/bark/assets/prompts/tr_speaker_4.npz differ diff --git a/bark/assets/prompts/tr_speaker_5.npz b/bark/assets/prompts/tr_speaker_5.npz new file mode 100644 index 0000000..adbbdec Binary files /dev/null and b/bark/assets/prompts/tr_speaker_5.npz differ diff --git a/bark/assets/prompts/tr_speaker_6.npz b/bark/assets/prompts/tr_speaker_6.npz new file mode 100644 index 0000000..2bc013e Binary files /dev/null and b/bark/assets/prompts/tr_speaker_6.npz differ diff --git a/bark/assets/prompts/tr_speaker_7.npz b/bark/assets/prompts/tr_speaker_7.npz new file mode 100644 index 0000000..c2154bc Binary files /dev/null and b/bark/assets/prompts/tr_speaker_7.npz differ diff --git a/bark/assets/prompts/tr_speaker_8.npz b/bark/assets/prompts/tr_speaker_8.npz new file mode 100644 index 0000000..09755e7 Binary files /dev/null and b/bark/assets/prompts/tr_speaker_8.npz differ diff --git a/bark/assets/prompts/tr_speaker_9.npz b/bark/assets/prompts/tr_speaker_9.npz new file mode 100644 index 0000000..9ccbd6e Binary files /dev/null and b/bark/assets/prompts/tr_speaker_9.npz differ diff --git a/bark/assets/prompts/zh_speaker_0.npz b/bark/assets/prompts/zh_speaker_0.npz new file mode 100644 index 0000000..c6181ab Binary files /dev/null and b/bark/assets/prompts/zh_speaker_0.npz differ diff --git a/bark/assets/prompts/zh_speaker_1.npz b/bark/assets/prompts/zh_speaker_1.npz new file mode 100644 index 0000000..8ce81fc Binary files /dev/null and b/bark/assets/prompts/zh_speaker_1.npz differ diff --git a/bark/assets/prompts/zh_speaker_2.npz b/bark/assets/prompts/zh_speaker_2.npz new file mode 100644 index 0000000..289fbb9 Binary files /dev/null and b/bark/assets/prompts/zh_speaker_2.npz differ diff --git a/bark/assets/prompts/zh_speaker_3.npz b/bark/assets/prompts/zh_speaker_3.npz new file mode 100644 index 0000000..02c352d Binary files /dev/null and b/bark/assets/prompts/zh_speaker_3.npz differ diff --git a/bark/assets/prompts/zh_speaker_4.npz b/bark/assets/prompts/zh_speaker_4.npz new file mode 100644 index 0000000..b696cf3 Binary files /dev/null and b/bark/assets/prompts/zh_speaker_4.npz differ diff --git a/bark/assets/prompts/zh_speaker_5.npz b/bark/assets/prompts/zh_speaker_5.npz new file mode 100644 index 0000000..5770a70 Binary files /dev/null and b/bark/assets/prompts/zh_speaker_5.npz differ diff --git a/bark/assets/prompts/zh_speaker_6.npz b/bark/assets/prompts/zh_speaker_6.npz new file mode 100644 index 0000000..1fa0e65 Binary files /dev/null and b/bark/assets/prompts/zh_speaker_6.npz differ diff --git a/bark/assets/prompts/zh_speaker_7.npz b/bark/assets/prompts/zh_speaker_7.npz new file mode 100644 index 0000000..50336bd Binary files /dev/null and b/bark/assets/prompts/zh_speaker_7.npz differ diff --git a/bark/assets/prompts/zh_speaker_8.npz b/bark/assets/prompts/zh_speaker_8.npz new file mode 100644 index 0000000..44d2218 Binary files /dev/null and b/bark/assets/prompts/zh_speaker_8.npz differ diff --git a/bark/assets/prompts/zh_speaker_9.npz b/bark/assets/prompts/zh_speaker_9.npz new file mode 100644 index 0000000..9d929ba Binary files /dev/null and b/bark/assets/prompts/zh_speaker_9.npz differ diff --git a/bark/generation.py b/bark/generation.py index 31aa39f..2b4fabe 100644 --- a/bark/generation.py +++ b/bark/generation.py @@ -48,10 +48,26 @@ COARSE_RATE_HZ = 75 SAMPLE_RATE = 24_000 -ALLOWED_PROMPTS = set( - [f"speech_{n}" for n in range(8)] + - [f"music_{n}" for n in range(6)] -) +SUPPORTED_LANGS = [ + ("English", "en"), + ("German", "de"), + ("Spanish", "es"), + ("French", "fr"), + ("Hindi", "hi"), + ("Italian", "it"), + ("Japanese", "ja"), + ("Korean", "ko"), + ("Polish", "pl"), + ("Portuguese", "pt"), + ("Russian", "ru"), + ("Turkish", "tr"), + ("Chinese", "zh"), +] + +ALLOWED_PROMPTS = {"announcer"} +for _, lang in SUPPORTED_LANGS: + for n in range(10): + ALLOWED_PROMPTS.add(f"{lang}_speaker_{n}") logger = logging.getLogger(__name__) @@ -83,6 +99,13 @@ REMOTE_MODEL_PATHS = { } +if not hasattr(torch.nn.functional, 'scaled_dot_product_attention'): + logger.warning( + "torch version does not support flash attention. You will get significantly faster" + + " inference speed by upgrade torch to newest version / nightly." + ) + + def _string_md5(s): m = hashlib.md5() m.update(s.encode("utf-8")) diff --git a/bark/notebooks/fake_classifier.ipynb b/notebooks/fake_classifier.ipynb similarity index 100% rename from bark/notebooks/fake_classifier.ipynb rename to notebooks/fake_classifier.ipynb