vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

tts : add OuteTTS support (#10784)

Browse source 
* server : add "tokens" output

ggml-ci

* server : output embeddings for all tokens when pooling = none

ggml-ci

* server : be explicit about the pooling type in the tests

ggml-ci

* server : do not normalize embeddings when there is no pooling

ggml-ci

* llama : add OuteTTS support (wip)

* wip

* extract features

* first conv

* group norm

* resnet conv

* resnet

* attn

* pos net

* layer norm

* convnext

* head

* hann window

* fix n_embd + remove llama.cpp hacks

* compute hann window

* fft

* spectrum processing

* clean-up

* tts : receive input text and generate codes

* clip : fix new conv name

* tts : minor fix

* tts : add header + minor fixes

ggml-ci

* tts : add matchematical constant

ggml-ci

* tts : fix sampling + cut initial noise

* tts : fixes

* tts : update default samplers

ggml-ci

* tts : text pre-processing

* tts : outetts-voc -> wavtokenizer-dec

* tts : remove hardcoded constants

ggml-ci

* tts : fix tensor shapes

* llama : refactor wavtokenizer tensors

ggml-ci

* cont

ggml-ci

* cont [no ci]

* llama : update WavTokenizer to non-causal attn

* llama : handle no-vocab detokenization

* tts : add Python example for OuteTTS (wip)

* tts : extend python example to generate spectrogram

ggml-ci

* server : fix rebase artifacts

* tts : enable "return_tokens" in Python example

ggml-ci

* tts : minor fixes

* common : support HF download for vocoder
This commit is contained in:
Georgi Gerganov 2024-12-18 19:27:21 +02:00 • committed by GitHub
parent 7bbb5acf12
commit 0bf2d10c55
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
19 changed files with 2509 additions and 532 deletions
Download patch file Download diff file Expand all files Collapse all files

21
gguf-py/gguf/gguf_writer.py
View file

@ -631,6 +631,21 @@ class GGUFWriter:
def add_embedding_length(self, length: int) -> None:
self.add_uint32(Keys.LLM.EMBEDDING_LENGTH.format(arch=self.arch), length)
def add_features_length(self, length: int) -> None:
self.add_uint32(Keys.LLM.FEATURES_LENGTH.format(arch=self.arch), length)
def add_posnet_embedding_length(self, length: int) -> None:
self.add_uint32(Keys.PosNet.EMBEDDING_LENGTH.format(arch=self.arch), length)
def add_posnet_block_count(self, length: int) -> None:
self.add_uint32(Keys.PosNet.BLOCK_COUNT.format(arch=self.arch), length)
def add_convnext_embedding_length(self, length: int) -> None:
self.add_uint32(Keys.ConvNext.EMBEDDING_LENGTH.format(arch=self.arch), length)
def add_convnext_block_count(self, length: int) -> None:
self.add_uint32(Keys.ConvNext.BLOCK_COUNT.format(arch=self.arch), length)
def add_block_count(self, length: int) -> None:
self.add_uint32(Keys.LLM.BLOCK_COUNT.format(arch=self.arch), length)
@ -727,6 +742,12 @@ class GGUFWriter:
def add_layer_norm_rms_eps(self, value: float) -> None:
self.add_float32(Keys.Attention.LAYERNORM_RMS_EPS.format(arch=self.arch), value)
def add_group_norm_eps(self, value: float) -> None:
self.add_float32(Keys.Attention.GROUPNORM_EPS.format(arch=self.arch), value)
def add_group_norm_groups(self, value: int) -> None:
self.add_uint32(Keys.Attention.GROUPNORM_GROUPS.format(arch=self.arch), value)
def add_causal_attention(self, value: bool) -> None:
self.add_bool(Keys.Attention.CAUSAL.format(arch=self.arch), value)