Clarifications regarding voices + fix missing @classmethod (#86)

This commit is contained in:
Václav Volhejn
2026-01-26 12:46:44 +01:00
committed by GitHub
parent edec7f8790
commit 77cc74a443
3 changed files with 21 additions and 0 deletions
+1
View File
@@ -61,6 +61,7 @@ for each voice.
The `--voice` argument can also take a plain wav file as input for voice cloning.
You can use your own or check out our [voice repository](https://huggingface.co/kyutai/tts-voices).
We recommend [cleaning the sample](https://podcast.adobe.com/en/enhance) before using it with Pocket TTS, because the audio quality of the sample is also reproduced.
Feel free to check out the [generate documentation](https://github.com/kyutai-labs/pocket-tts/tree/main/docs/generate.md) for more details and examples.
For trying multiple voices and prompts quickly, prefer using the `serve` command.
+2
View File
@@ -173,7 +173,9 @@ class TTSModel(nn.Module):
return tts_model
@classmethod
def load_model(
cls,
variant: str = DEFAULT_VARIANT,
temp: float | int = DEFAULT_TEMPERATURE,
lsd_decode_steps: int = DEFAULT_LSD_DECODE_STEPS,
+18
View File
@@ -46,6 +46,15 @@
Supports: http://, https://, or hf:// URLs.<br>
You can also use predefined voices:<br>
"alba", "marius", "javert", "jean", "fantine", "cosette", "eponine", "azelma".
<br/>
You can find more voices in our
<a
href="https://huggingface.co/kyutai/tts-voices"
target="_blank"
rel="noopener noreferrer"
class="text-blue-600 hover:underline"
>
voice repository</a>.
</p>
</div>
@@ -61,6 +70,15 @@
/>
<p class="text-xs text-gray-500 mt-1">
Upload an audio file (WAV, MP3, FLAC, etc.) to use as voice reference. Takes precedence over voice URL.
For best results,
<a
href="https://podcast.adobe.com/en/enhance"
target="_blank"
rel="noopener noreferrer"
class="text-blue-600 hover:underline"
>clean the sample</a
>
first.
</p>
</div>