diff --git a/cloned-voices/aria.wav b/cloned-voices/aria.wav new file mode 100644 index 0000000..e3826f9 Binary files /dev/null and b/cloned-voices/aria.wav differ diff --git a/cloned-voices/atlas.wav b/cloned-voices/atlas.wav new file mode 100644 index 0000000..bfe2331 Binary files /dev/null and b/cloned-voices/atlas.wav differ diff --git a/cloned-voices/caleb.wav b/cloned-voices/caleb.wav new file mode 100644 index 0000000..03b25b7 Binary files /dev/null and b/cloned-voices/caleb.wav differ diff --git a/cloned-voices/clara.wav b/cloned-voices/clara.wav new file mode 100644 index 0000000..a75a4eb Binary files /dev/null and b/cloned-voices/clara.wav differ diff --git a/cloned-voices/elena.wav b/cloned-voices/elena.wav new file mode 100644 index 0000000..6030f75 Binary files /dev/null and b/cloned-voices/elena.wav differ diff --git a/cloned-voices/felix.wav b/cloned-voices/felix.wav new file mode 100644 index 0000000..029f11c Binary files /dev/null and b/cloned-voices/felix.wav differ diff --git a/cloned-voices/grace.wav b/cloned-voices/grace.wav new file mode 100644 index 0000000..4d9f182 Binary files /dev/null and b/cloned-voices/grace.wav differ diff --git a/cloned-voices/hazel.wav b/cloned-voices/hazel.wav new file mode 100644 index 0000000..8685a71 Binary files /dev/null and b/cloned-voices/hazel.wav differ diff --git a/cloned-voices/hugo.wav b/cloned-voices/hugo.wav new file mode 100644 index 0000000..1b2adae Binary files /dev/null and b/cloned-voices/hugo.wav differ diff --git a/cloned-voices/iris.wav b/cloned-voices/iris.wav new file mode 100644 index 0000000..544b507 Binary files /dev/null and b/cloned-voices/iris.wav differ diff --git a/cloned-voices/jasper.wav b/cloned-voices/jasper.wav new file mode 100644 index 0000000..0ed3c92 Binary files /dev/null and b/cloned-voices/jasper.wav differ diff --git a/cloned-voices/kai.wav b/cloned-voices/kai.wav new file mode 100644 index 0000000..6725f43 Binary files /dev/null and b/cloned-voices/kai.wav differ diff --git a/cloned-voices/leo.wav b/cloned-voices/leo.wav new file mode 100644 index 0000000..b911ecb Binary files /dev/null and b/cloned-voices/leo.wav differ diff --git a/cloned-voices/luna.wav b/cloned-voices/luna.wav new file mode 100644 index 0000000..34310da Binary files /dev/null and b/cloned-voices/luna.wav differ diff --git a/cloned-voices/marcus.wav b/cloned-voices/marcus.wav new file mode 100644 index 0000000..df15a26 Binary files /dev/null and b/cloned-voices/marcus.wav differ diff --git a/cloned-voices/maya.wav b/cloned-voices/maya.wav new file mode 100644 index 0000000..cdab383 Binary files /dev/null and b/cloned-voices/maya.wav differ diff --git a/cloned-voices/owen.wav b/cloned-voices/owen.wav new file mode 100644 index 0000000..915039d Binary files /dev/null and b/cloned-voices/owen.wav differ diff --git a/cloned-voices/ruby.wav b/cloned-voices/ruby.wav new file mode 100644 index 0000000..fe8256c Binary files /dev/null and b/cloned-voices/ruby.wav differ diff --git a/cloned-voices/sage.wav b/cloned-voices/sage.wav new file mode 100644 index 0000000..9ad7aca Binary files /dev/null and b/cloned-voices/sage.wav differ diff --git a/cloned-voices/sofia.wav b/cloned-voices/sofia.wav new file mode 100644 index 0000000..a9c629b Binary files /dev/null and b/cloned-voices/sofia.wav differ diff --git a/cloned-voices/theo.wav b/cloned-voices/theo.wav new file mode 100644 index 0000000..363b16b Binary files /dev/null and b/cloned-voices/theo.wav differ diff --git a/cloned-voices/voices_metadata.json b/cloned-voices/voices_metadata.json new file mode 100644 index 0000000..2a1142e --- /dev/null +++ b/cloned-voices/voices_metadata.json @@ -0,0 +1,170 @@ +{ + "aria": { + "ref_audio": "cloned-voices/aria.wav", + "ref_text": "THE PRIDE OF THAT DIM IMAGE BROUGHT BACK TO HIS MIND THE DIGNITY OF THE OFFICE HE HAD REFUSED", + "speaker_id": "1089", + "gender": "female", + "description": "neutral, balanced - clear professional delivery", + "duration": 5.9 + }, + "clara": { + "ref_audio": "cloned-voices/clara.wav", + "ref_text": "BUT IN THIS VIGNETTE COPIED FROM TURNER YOU HAVE THE TWO PRINCIPLES BROUGHT OUT PERFECTLY", + "speaker_id": "1188", + "gender": "female", + "description": "warm, friendly - approachable delivery", + "duration": 6.1 + }, + "elena": { + "ref_audio": "cloned-voices/elena.wav", + "ref_text": "YEA HIS HONOURABLE WORSHIP IS WITHIN BUT HE HATH A GODLY MINISTER OR TWO WITH HIM AND LIKEWISE A LEECH", + "speaker_id": "1221", + "gender": "female", + "description": "expressive, storyteller - engaging narration", + "duration": 7.1 + }, + "grace": { + "ref_audio": "cloned-voices/grace.wav", + "ref_text": "FOUR OR FIVE OF THE LATTER ONLY LINGERED ABOUT THE DOOR OF THE PRISON OF UNCAS WARY BUT CLOSE OBSERVERS OF THE MANNER OF THEIR CAPTIVE", + "speaker_id": "1320", + "gender": "female", + "description": "soft, gentle - calm and soothing", + "duration": 7.6 + }, + "hazel": { + "ref_audio": "cloned-voices/hazel.wav", + "ref_text": "BUT IT IS NOT WITH A VIEW TO DISTINCTION THAT YOU SHOULD CULTIVATE THIS TALENT IF YOU CONSULT YOUR OWN HAPPINESS", + "speaker_id": "3575", + "gender": "female", + "description": "calm, neutral - steady and even", + "duration": 7.0 + }, + "iris": { + "ref_audio": "cloned-voices/iris.wav", + "ref_text": "NOW WHEN HAS HORROR EVER EXCLUDED STUDY", + "speaker_id": "4507", + "gender": "female", + "description": "young, clear - crisp enunciation", + "duration": 6.1 + }, + "luna": { + "ref_audio": "cloned-voices/luna.wav", + "ref_text": "THE WEARERS OF UNIFORMS AND LIVERIES MAY BE ROUGHLY DIVIDED INTO TWO CLASSES THE FREE AND THE SERVILE OR THE NOBLE AND THE IGNOBLE", + "speaker_id": "3570", + "gender": "female", + "description": "bright, energetic - lively and upbeat", + "duration": 7.8 + }, + "maya": { + "ref_audio": "cloned-voices/maya.wav", + "ref_text": "THE DOOR OPENED AGAIN WHILE I WAS STILL STUDYING THE TWO BROTHERS WITHOUT I HONESTLY CONFESS BEING VERY FAVORABLY IMPRESSED BY EITHER OF THEM", + "speaker_id": "5142", + "gender": "female", + "description": "warm, conversational - natural flow", + "duration": 7.1 + }, + "ruby": { + "ref_audio": "cloned-voices/ruby.wav", + "ref_text": "HE HAD PRECONCEIVED IDEAS ABOUT EVERYTHING AND HIS IDEA ABOUT AMERICANS WAS THAT THEY SHOULD BE ENGINEERS OR MECHANICS", + "speaker_id": "4446", + "gender": "female", + "description": "mature, authoritative - confident tone", + "duration": 6.3 + }, + "sage": { + "ref_audio": "cloned-voices/sage.wav", + "ref_text": "MISS DE GRAF SAID KENNETH NOTICING THE BOY'S FACE CRITICALLY AS HE STOOD WHERE THE LIGHT FROM THE PASSAGE FELL UPON IT", + "speaker_id": "6829", + "gender": "female", + "description": "crisp, articulate - precise diction", + "duration": 7.1 + }, + "sofia": { + "ref_audio": "cloned-voices/sofia.wav", + "ref_text": "IT IS YOU WHO ARE MISTAKEN RAOUL I HAVE READ HIS DISTRESS IN HIS EYES IN HIS EVERY GESTURE AND ACTION THE WHOLE DAY", + "speaker_id": "6930", + "gender": "female", + "description": "smooth, polished - radio-style delivery", + "duration": 7.3 + }, + "atlas": { + "ref_audio": "cloned-voices/atlas.wav", + "ref_text": "SHE SAW THAT THE BED WAS GILDED AND SO RICH THAT IT SEEMED THAT OF A PRINCE RATHER THAN OF A PRIVATE GENTLEMAN", + "speaker_id": "5639", + "gender": "male", + "description": "deep, dramatic - authoritative tone", + "duration": 6.9 + }, + "caleb": { + "ref_audio": "cloned-voices/caleb.wav", + "ref_text": "GOLD IS THE MOST COMMON METAL IN THE LAND OF OZ AND IS USED FOR MANY PURPOSES BECAUSE IT IS SOFT AND PLIABLE", + "speaker_id": "1284", + "gender": "male", + "description": "clear, measured - precise enunciation", + "duration": 7.2 + }, + "felix": { + "ref_audio": "cloned-voices/felix.wav", + "ref_text": "THEN TURNING TO JANE SHE ASKED IN A SOMEWHAT ALTERED TONE HAS SHE BEEN A GOOD GIRL JANE", + "speaker_id": "7021", + "gender": "male", + "description": "light, conversational - casual tone", + "duration": 7.0 + }, + "hugo": { + "ref_audio": "cloned-voices/hugo.wav", + "ref_text": "IN THE OLD BADLY MADE PLAY IT WAS FREQUENTLY NECESSARY FOR ONE OF THE CHARACTERS TO TAKE THE AUDIENCE INTO HIS CONFIDENCE", + "speaker_id": "7176", + "gender": "male", + "description": "warm storyteller - engaging narrative", + "duration": 7.2 + }, + "jasper": { + "ref_audio": "cloned-voices/jasper.wav", + "ref_text": "AT THE INCEPTION OF PLURAL MARRIAGE AMONG THE LATTER DAY SAINTS THERE WAS NO LAW NATIONAL OR STATE AGAINST ITS PRACTISE", + "speaker_id": "4077", + "gender": "male", + "description": "mature, professor - thoughtful pacing", + "duration": 7.7 + }, + "kai": { + "ref_audio": "cloned-voices/kai.wav", + "ref_text": "I KNEW NOTHING OF THE DOCTRINE OF FAITH BECAUSE WE WERE TAUGHT SOPHISTRY INSTEAD OF CERTAINTY AND NOBODY UNDERSTOOD SPIRITUAL BOASTING", + "speaker_id": "2830", + "gender": "male", + "description": "young, energetic - dynamic delivery", + "duration": 7.0 + }, + "leo": { + "ref_audio": "cloned-voices/leo.wav", + "ref_text": "RUTH SAT QUITE STILL FOR A TIME WITH FACE INTENT AND FLUSHED IT WAS OUT NOW", + "speaker_id": "4970", + "gender": "male", + "description": "smooth announcer - polished delivery", + "duration": 6.9 + }, + "marcus": { + "ref_audio": "cloned-voices/marcus.wav", + "ref_text": "BUT THE WINDOWS ARE PATCHED WITH WOODEN PANES AND THE DOOR I THINK IS LIKE THE GATE IT IS NEVER OPENED", + "speaker_id": "2094", + "gender": "male", + "description": "warm baritone - rich and resonant", + "duration": 8.0 + }, + "owen": { + "ref_audio": "cloned-voices/owen.wav", + "ref_text": "AND WHAT DEMONSTRATION DO YOU OFFER ASKED SERVADAC EAGERLY THAT IT WILL NOT HAPPEN", + "speaker_id": "5105", + "gender": "male", + "description": "neutral, clear - straightforward style", + "duration": 7.0 + }, + "theo": { + "ref_audio": "cloned-voices/theo.wav", + "ref_text": "THE ALTERNATIVE WAS THAT SOMEONE PASSING HAD OBSERVED THE KEY IN THE DOOR HAD KNOWN THAT I WAS OUT AND HAD ENTERED TO LOOK AT THE PAPERS", + "speaker_id": "1580", + "gender": "male", + "description": "clear narrator - smooth reading style", + "duration": 7.0 + } +} \ No newline at end of file diff --git a/voice_to_speaker.default.yaml b/voice_to_speaker.default.yaml index 7217389..134f8bf 100644 --- a/voice_to_speaker.default.yaml +++ b/voice_to_speaker.default.yaml @@ -1,133 +1,134 @@ # uncloseai-speech Voice Configuration -# Diverse voice samples for Qwen3-TTS voice cloning +# Diverse voice samples from LibriSpeech test-clean (public domain) +# Each voice is a DISTINCT SPEAKER for Qwen3-TTS voice cloning # # Standard voices: alloy, echo, fable, onyx, nova, shimmer # Extended voices: amber, breeze, coral, dawn, ember, frost, # glow, haze, ivy, jade, kite, lark, mist, nectar # -# Source: LJ Speech Dataset (public domain) -# https://keithito.com/LJ-Speech-Dataset/ +# Source: LibriSpeech test-clean (public domain, LibriVox recordings) +# 20 distinct speakers (10 female, 10 male) tts-1-qwen: # neutral, balanced - clear professional delivery alloy: ref_audio: cloned-voices/alloy.wav - ref_text: "Printing, in the only sense with which we are at present concerned, differs from most if not from all the arts and crafts represented in the Exhibition" + ref_text: "THE PRIDE OF THAT DIM IMAGE BROUGHT BACK TO HIS MIND THE DIGNITY OF THE OFFICE HE HAD REFUSED" language: English # clear, measured - precise enunciation echo: ref_audio: cloned-voices/echo.wav - ref_text: "yet the art did not begin to flourish in Europe until the middle of the fifteenth century." + ref_text: "GOLD IS THE MOST COMMON METAL IN THE LAND OF OZ AND IS USED FOR MANY PURPOSES BECAUSE IT IS SOFT AND PLIABLE" language: English # expressive, storyteller - engaging narration fable: ref_audio: cloned-voices/fable.wav - ref_text: "The type of this time in spite of the many failures is in the main admirable" + ref_text: "YEA HIS HONOURABLE WORSHIP IS WITHIN BUT HE HATH A GODLY MINISTER OR TWO WITH HIM AND LIKEWISE A LEECH" language: English # deep, dramatic - authoritative tone onyx: ref_audio: cloned-voices/onyx.wav - ref_text: "For although the Chinese seem to have known its art of printing with engraved wooden blocks" + ref_text: "SHE SAW THAT THE BED WAS GILDED AND SO RICH THAT IT SEEMED THAT OF A PRINCE RATHER THAN OF A PRIVATE GENTLEMAN" language: English # warm, friendly - approachable delivery nova: ref_audio: cloned-voices/nova.wav - ref_text: "the art of block printing was known in Europe during the first half of the fifteenth century" + ref_text: "BUT IN THIS VIGNETTE COPIED FROM TURNER YOU HAVE THE TWO PRINCIPLES BROUGHT OUT PERFECTLY" language: English # soft, gentle - calm and soothing shimmer: ref_audio: cloned-voices/shimmer.wav - ref_text: "in being comparatively modern." + ref_text: "FOUR OR FIVE OF THE LATTER ONLY LINGERED ABOUT THE DOOR OF THE PRISON OF UNCAS WARY BUT CLOSE OBSERVERS OF THE MANNER OF THEIR CAPTIVE" language: English - # warm amber glow - inviting and comfortable + # warm, conversational - natural flow amber: ref_audio: cloned-voices/amber.wav - ref_text: "it may be necessary to turn over many examples before finding one that is even passable" + ref_text: "THE DOOR OPENED AGAIN WHILE I WAS STILL STUDYING THE TWO BROTHERS WITHOUT I HONESTLY CONFESS BEING VERY FAVORABLY IMPRESSED BY EITHER OF THEM" language: English - # light breeze - airy and refreshing + # bright, energetic - lively and upbeat breeze: ref_audio: cloned-voices/breeze.wav - ref_text: "the commonest, that is to say, the most familiar faces, depart a good deal from those of the least common types" + ref_text: "THE WEARERS OF UNIFORMS AND LIVERIES MAY BE ROUGHLY DIVIDED INTO TWO CLASSES THE FREE AND THE SERVILE OR THE NOBLE AND THE IGNOBLE" language: English - # coral reef - vibrant and lively + # mature, authoritative - confident tone coral: ref_audio: cloned-voices/coral.wav - ref_text: "The practice of the earlier printers led them to attach the pieces of the text carefully together" + ref_text: "HE HAD PRECONCEIVED IDEAS ABOUT EVERYTHING AND HIS IDEA ABOUT AMERICANS WAS THAT THEY SHOULD BE ENGINEERS OR MECHANICS" language: English - # early dawn - fresh and hopeful + # crisp, articulate - precise diction dawn: ref_audio: cloned-voices/dawn.wav - ref_text: "This practice has spoiled many books from many different points of view" + ref_text: "MISS DE GRAF SAID KENNETH NOTICING THE BOY'S FACE CRITICALLY AS HE STOOD WHERE THE LIGHT FROM THE PASSAGE FELL UPON IT" language: English - # glowing ember - warm with depth - ember: - ref_audio: cloned-voices/ember.wav - ref_text: "Indeed, it is still the case that a good many examples of mediaeval printing" - language: English - - # winter frost - crisp and clear - frost: - ref_audio: cloned-voices/frost.wav - ref_text: "In spite of the many errors both of commission and omission in which the early printers" - language: English - - # soft glow - gentle radiance + # calm, neutral - steady and even glow: ref_audio: cloned-voices/glow.wav - ref_text: "as the types used are of necessity identical, it is obvious that for the sake of appearance" + ref_text: "BUT IT IS NOT WITH A VIEW TO DISTINCTION THAT YOU SHOULD CULTIVATE THIS TALENT IF YOU CONSULT YOUR OWN HAPPINESS" language: English - # morning haze - dreamy and ethereal - haze: - ref_audio: cloned-voices/haze.wav - ref_text: "the character of the letters forming a font" - language: English - - # climbing ivy - natural and organic + # young, clear - crisp enunciation ivy: ref_audio: cloned-voices/ivy.wav - ref_text: "which allows for the production of books of all degrees of excellence" + ref_text: "NOW WHEN HAS HORROR EVER EXCLUDED STUDY" language: English - # jade stone - smooth and precious + # warm baritone - rich and resonant + ember: + ref_audio: cloned-voices/ember.wav + ref_text: "BUT THE WINDOWS ARE PATCHED WITH WOODEN PANES AND THE DOOR I THINK IS LIKE THE GATE IT IS NEVER OPENED" + language: English + + # smooth announcer - polished delivery + frost: + ref_audio: cloned-voices/frost.wav + ref_text: "RUTH SAT QUITE STILL FOR A TIME WITH FACE INTENT AND FLUSHED IT WAS OUT NOW" + language: English + + # light, conversational - casual tone + haze: + ref_audio: cloned-voices/haze.wav + ref_text: "THEN TURNING TO JANE SHE ASKED IN A SOMEWHAT ALTERED TONE HAS SHE BEEN A GOOD GIRL JANE" + language: English + + # mature, professor - thoughtful pacing jade: ref_audio: cloned-voices/jade.wav - ref_text: "he must be able to draw his letter well and make his curves in geometry" + ref_text: "AT THE INCEPTION OF PLURAL MARRIAGE AMONG THE LATTER DAY SAINTS THERE WAS NO LAW NATIONAL OR STATE AGAINST ITS PRACTISE" language: English - # flying kite - free and playful + # young, energetic - dynamic delivery kite: ref_audio: cloned-voices/kite.wav - ref_text: "From time to time this subject has been much debated" + ref_text: "I KNEW NOTHING OF THE DOCTRINE OF FAITH BECAUSE WE WERE TAUGHT SOPHISTRY INSTEAD OF CERTAINTY AND NOBODY UNDERSTOOD SPIRITUAL BOASTING" language: English - # morning lark - cheerful and bright + # warm storyteller - engaging narrative lark: ref_audio: cloned-voices/lark.wav - ref_text: "Again, it is of the utmost importance that the types which we call the roman" + ref_text: "IN THE OLD BADLY MADE PLAY IT WAS FREQUENTLY NECESSARY FOR ONE OF THE CHARACTERS TO TAKE THE AUDIENCE INTO HIS CONFIDENCE" language: English - # soft mist - mysterious and intriguing + # neutral, clear - straightforward style mist: ref_audio: cloned-voices/mist.wav - ref_text: "Now, as all books not primarily intended as picture-books consist principally of types composed" + ref_text: "AND WHAT DEMONSTRATION DO YOU OFFER ASKED SERVADAC EAGERLY THAT IT WILL NOT HAPPEN" language: English - # sweet nectar - rich and delightful + # clear narrator - smooth reading style nectar: ref_audio: cloned-voices/nectar.wav - ref_text: "The other matter to be considered is the arrangement of the printed matter" + ref_text: "THE ALTERNATIVE WAS THAT SOMEONE PASSING HAD OBSERVED THE KEY IN THE DOOR HAD KNOWN THAT I WAS OUT AND HAD ENTERED TO LOOK AT THE PAPERS" language: English # Other TTS engines (disabled by default)