Reduce memory usage even more + better sampling

- The encode/decode memory buffers are now reused
- If the 30-sec segment goes for too long without a timestamp token, we
  force one. Improves transcription for large model
- Stereo support
- Add "micro-machines.wav" sample
pull/19/head
Georgi Gerganov 2022-09-30 19:33:09 +03:00
parent 310f4883d1
commit 3bcdbdfc32
No known key found for this signature in database
GPG Key ID: 449E073F9DC10735
3 changed files with 96 additions and 87 deletions

View File

@ -20,10 +20,13 @@ samples:
@wget --quiet --show-progress -O samples/gb0.ogg https://upload.wikimedia.org/wikipedia/commons/2/22/George_W._Bush%27s_weekly_radio_address_%28November_1%2C_2008%29.oga @wget --quiet --show-progress -O samples/gb0.ogg https://upload.wikimedia.org/wikipedia/commons/2/22/George_W._Bush%27s_weekly_radio_address_%28November_1%2C_2008%29.oga
@wget --quiet --show-progress -O samples/gb1.ogg https://upload.wikimedia.org/wikipedia/commons/1/1f/George_W_Bush_Columbia_FINAL.ogg @wget --quiet --show-progress -O samples/gb1.ogg https://upload.wikimedia.org/wikipedia/commons/1/1f/George_W_Bush_Columbia_FINAL.ogg
@wget --quiet --show-progress -O samples/hp0.ogg https://upload.wikimedia.org/wikipedia/en/d/d4/En.henryfphillips.ogg @wget --quiet --show-progress -O samples/hp0.ogg https://upload.wikimedia.org/wikipedia/en/d/d4/En.henryfphillips.ogg
@wget --quiet --show-progress -O samples/mm1.wav https://cdn.openai.com/whisper/draft-20220913a/micro-machines.wav
@echo "Converting to 16-bit WAV ..." @echo "Converting to 16-bit WAV ..."
@ffmpeg -loglevel -0 -y -i samples/gb0.ogg -ar 16000 -ac 1 -c:a pcm_s16le samples/gb0.wav @ffmpeg -loglevel -0 -y -i samples/gb0.ogg -ar 16000 -ac 1 -c:a pcm_s16le samples/gb0.wav
@ffmpeg -loglevel -0 -y -i samples/gb1.ogg -ar 16000 -ac 1 -c:a pcm_s16le samples/gb1.wav @ffmpeg -loglevel -0 -y -i samples/gb1.ogg -ar 16000 -ac 1 -c:a pcm_s16le samples/gb1.wav
@ffmpeg -loglevel -0 -y -i samples/hp0.ogg -ar 16000 -ac 1 -c:a pcm_s16le samples/hp0.wav @ffmpeg -loglevel -0 -y -i samples/hp0.ogg -ar 16000 -ac 1 -c:a pcm_s16le samples/hp0.wav
@ffmpeg -loglevel -0 -y -i samples/mm1.wav -ar 16000 -ac 1 -c:a pcm_s16le samples/mm0.wav
@rm samples/mm1.wav
# if not already downloaded, the following targets download the specified model and # if not already downloaded, the following targets download the specified model and

View File

@ -1,12 +1,13 @@
# whisper.cpp # whisper.cpp
C/C++ port of [OpenAI's Whisper](https://github.com/openai/whisper) speech-to-text model High-performance inference of [OpenAI's Whisper](https://github.com/openai/whisper) automatic speech recognition (ASR) model:
- Plain C/C++ implementation without dependencies - Plain C/C++ implementation without dependencies
- ARM_NEON and AVX intrinsics support - ARM_NEON and AVX intrinsics support
- Mixed F16 / F32 support - Mixed F16 / F32 support
- Low memory usage (Flash Attention + Flash Forward) - Low memory usage (Flash Attention + Flash Forward)
- Zero memory allocations at runtime - Zero memory allocations at runtime
- Runs on the CPU (Mac and Linux support)
## Usage ## Usage
@ -50,7 +51,12 @@ options:
bash ./download-ggml-model.sh base.en bash ./download-ggml-model.sh base.en
Downloading ggml model base.en ... Downloading ggml model base.en ...
Model base.en already exists. Skipping download. models/ggml-base.en.bin 100%[=====================================>] 141.11M 8.58MB/s in 22s
Done! Model 'base.en' saved in 'models/ggml-base.en.bin'
You can now use it like this:
$ ./main -m models/ggml-base.en.bin -f samples/jfk.wav
=============================================== ===============================================
Running base.en on all samples in ./samples ... Running base.en on all samples in ./samples ...
@ -73,7 +79,7 @@ whisper_model_load: n_text_layer = 6
whisper_model_load: n_mels = 80 whisper_model_load: n_mels = 80
whisper_model_load: f16 = 1 whisper_model_load: f16 = 1
whisper_model_load: type = 2 whisper_model_load: type = 2
whisper_model_load: mem_required = 611.00 MB whisper_model_load: mem_required = 377.00 MB
whisper_model_load: adding 1607 extra tokens whisper_model_load: adding 1607 extra tokens
whisper_model_load: ggml ctx size = 163.43 MB whisper_model_load: ggml ctx size = 163.43 MB
whisper_model_load: memory size = 22.83 MB whisper_model_load: memory size = 22.83 MB
@ -86,12 +92,12 @@ main: processing 176000 samples (11.0 sec), 4 threads, lang = english, task = tr
[00:00.000 --> 00:11.000] And so my fellow Americans ask not what your country can do for you. Ask what you can do for your country. [00:00.000 --> 00:11.000] And so my fellow Americans ask not what your country can do for you. Ask what you can do for your country.
main: load time = 61.78 ms main: load time = 82.05 ms
main: mel time = 41.74 ms main: mel time = 44.15 ms
main: sample time = 2.10 ms main: sample time = 1.98 ms
main: encode time = 718.60 ms / 119.77 ms per layer main: encode time = 674.77 ms / 112.46 ms per layer
main: decode time = 83.55 ms main: decode time = 82.91 ms
main: total time = 908.15 ms main: total time = 886.29 ms
``` ```
The command downloads the `base.en` model converted to custom `ggml` format and runs the inference on all `.wav` samples in the folder `samples`. The command downloads the `base.en` model converted to custom `ggml` format and runs the inference on all `.wav` samples in the folder `samples`.
@ -131,10 +137,12 @@ make large
## Another example ## Another example
Here is another example of transcribing a [3:24 min speech](https://upload.wikimedia.org/wikipedia/commons/1/1f/George_W_Bush_Columbia_FINAL.ogg) in less than a minute, using `medium.en` model: Here is another example of transcribing a [3:24 min speech](https://upload.wikimedia.org/wikipedia/commons/1/1f/George_W_Bush_Columbia_FINAL.ogg)
in less than a minute on a MacBook M1 Pro, using `medium.en` model:
```java ```java
$ ./main -m models/ggml-medium.en.bin -f samples/gb1.wav -t 8 $ ./main -m models/ggml-medium.en.bin -f samples/gb1.wav -t 8
whisper_model_load: loading model from 'models/ggml-medium.en.bin' whisper_model_load: loading model from 'models/ggml-medium.en.bin'
whisper_model_load: n_vocab = 51864 whisper_model_load: n_vocab = 51864
whisper_model_load: n_audio_ctx = 1500 whisper_model_load: n_audio_ctx = 1500
@ -148,7 +156,7 @@ whisper_model_load: n_text_layer = 24
whisper_model_load: n_mels = 80 whisper_model_load: n_mels = 80
whisper_model_load: f16 = 1 whisper_model_load: f16 = 1
whisper_model_load: type = 4 whisper_model_load: type = 4
whisper_model_load: mem_required = 2786.00 MB whisper_model_load: mem_required = 2502.00 MB
whisper_model_load: adding 1607 extra tokens whisper_model_load: adding 1607 extra tokens
whisper_model_load: ggml ctx size = 1644.97 MB whisper_model_load: ggml ctx size = 1644.97 MB
whisper_model_load: memory size = 182.62 MB whisper_model_load: memory size = 182.62 MB
@ -187,30 +195,30 @@ main: processing 3179750 samples (198.7 sec), 8 threads, lang = english, task =
[03:14.000 --> 03:24.000] [Music] [03:14.000 --> 03:24.000] [Music]
main: load time = 438.55 ms main: load time = 522.18 ms
main: mel time = 440.22 ms main: mel time = 423.43 ms
main: sample time = 32.23 ms main: sample time = 31.42 ms
main: encode time = 42329.63 ms / 1763.73 ms per layer main: encode time = 41518.51 ms / 1729.94 ms per layer
main: decode time = 15190.00 ms main: decode time = 14907.22 ms
main: total time = 58444.63 ms main: total time = 57416.63 ms
``` ```
## Limitations ## Limitations
- Very basic greedy sampling scheme - always pick up the top token - Very basic greedy sampling scheme - always pick up the top token
- Only 16-bit WAV at 16 kHz is supported
- Inference only - Inference only
- Runs on the CPU - No GPU support
- Only mono-channel 16-bit WAV is supported
## Memory usage ## Memory usage
| Model | Disk | Mem | | Model | Disk | Mem |
| --- | --- | --- | | --- | --- | --- |
| tiny | 75 MB | ~460 MB | | tiny | 75 MB | ~240 MB |
| base | 142 MB | ~620 MB | | base | 142 MB | ~380 MB |
| small | 466 MB | ~1.3 GB | | small | 466 MB | ~970 MB |
| medium | 1.5 GB | ~2.8 GB | | medium | 1.5 GB | ~2.5 GB |
| large | 2.9 GB | ~4.9 GB | | large | 2.9 GB | ~4.6 GB |
## ggml format ## ggml format

120
main.cpp
View File

@ -158,11 +158,11 @@ const std::map<e_model, size_t> MEM_REQ_ENCODE_LAYER = {
}; };
const std::map<e_model, size_t> MEM_REQ_DECODE = { const std::map<e_model, size_t> MEM_REQ_DECODE = {
{ MODEL_TINY, 190ull*MB }, { MODEL_TINY, 94ull*MB },
{ MODEL_BASE, 190ull*MB }, { MODEL_BASE, 96ull*MB },
{ MODEL_SMALL, 190ull*MB }, { MODEL_SMALL, 98ull*MB },
{ MODEL_MEDIUM, 200ull*MB }, { MODEL_MEDIUM, 100ull*MB },
{ MODEL_LARGE, 200ull*MB }, { MODEL_LARGE, 102ull*MB },
}; };
const std::map<e_model, size_t> MEM_REQ_DECODE_LAYER = { const std::map<e_model, size_t> MEM_REQ_DECODE_LAYER = {
@ -173,6 +173,11 @@ const std::map<e_model, size_t> MEM_REQ_DECODE_LAYER = {
{ MODEL_LARGE, 110ull*MB }, { MODEL_LARGE, 110ull*MB },
}; };
// the memory buffers used to store the model in memory and perform the inference computations
std::vector<uint8_t> g_buf_model;
std::vector<uint8_t> g_buf_compute;
std::vector<uint8_t> g_buf_compute_layer;
const int SAMPLE_RATE = 16000; const int SAMPLE_RATE = 16000;
const int N_FFT = 400; const int N_FFT = 400;
const int N_MEL = 80; const int N_MEL = 80;
@ -542,13 +547,15 @@ bool whisper_model_load(const std::string & fname, whisper_model & model, whispe
printf("%s: f16 = %d\n", __func__, hparams.f16); printf("%s: f16 = %d\n", __func__, hparams.f16);
printf("%s: type = %d\n", __func__, model.type); printf("%s: type = %d\n", __func__, model.type);
g_buf_model.resize(MEM_REQ_MODEL.at(model.type));
g_buf_compute.resize(std::max(MEM_REQ_ENCODE.at(model.type), MEM_REQ_DECODE.at(model.type)));
g_buf_compute_layer.resize(std::max(MEM_REQ_ENCODE_LAYER.at(model.type), MEM_REQ_DECODE_LAYER.at(model.type)));
// this is the total memory required to run the inference // this is the total memory required to run the inference
const size_t mem_required = const size_t mem_required =
MEM_REQ_MODEL.at(model.type) + g_buf_model.size() +
MEM_REQ_ENCODE.at(model.type) + g_buf_compute.size() +
MEM_REQ_ENCODE_LAYER.at(model.type) + g_buf_compute_layer.size();
MEM_REQ_DECODE.at(model.type) +
MEM_REQ_DECODE_LAYER.at(model.type);
printf("%s: mem_required = %.2f MB\n", __func__, mem_required / 1024.0 / 1024.0); printf("%s: mem_required = %.2f MB\n", __func__, mem_required / 1024.0 / 1024.0);
} }
@ -752,8 +759,8 @@ bool whisper_model_load(const std::string & fname, whisper_model & model, whispe
// create the ggml context // create the ggml context
{ {
struct ggml_init_params params = { struct ggml_init_params params = {
.mem_size = ctx_size, .mem_size = g_buf_model.size(),
.mem_buffer = NULL, .mem_buffer = g_buf_model.data(),
}; };
model.ctx = ggml_init(params); model.ctx = ggml_init(params);
@ -1089,17 +1096,10 @@ bool whisper_encode(
const int n_mels = hparams.n_mels; const int n_mels = hparams.n_mels;
assert(mel_inp.n_mel == n_mels); assert(mel_inp.n_mel == n_mels);
struct ggml_init_params params; struct ggml_init_params params = {
.mem_size = g_buf_compute.size(),
{ .mem_buffer = g_buf_compute.data(),
static size_t buf_size = MEM_REQ_ENCODE.at(model.type); };
static void * buf = malloc(buf_size);
params = {
.mem_size = buf_size,
.mem_buffer = buf,
};
}
struct ggml_context * ctx0 = ggml_init(params); struct ggml_context * ctx0 = ggml_init(params);
@ -1151,16 +1151,10 @@ bool whisper_encode(
// create separate context for each layer to reduce memory usage // create separate context for each layer to reduce memory usage
struct ggml_init_params paramsL; struct ggml_init_params paramsL = {
{ .mem_size = g_buf_compute_layer.size(),
static size_t buf_size = MEM_REQ_ENCODE_LAYER.at(model.type); .mem_buffer = g_buf_compute_layer.data(),
static void * buf = malloc(buf_size); };
paramsL = {
.mem_size = buf_size,
.mem_buffer = buf,
};
}
struct ggml_context * ctxL = ggml_init(paramsL); struct ggml_context * ctxL = ggml_init(paramsL);
@ -1492,17 +1486,10 @@ bool whisper_decode(
const int N = prompt.size(); const int N = prompt.size();
const int M = hparams.n_audio_ctx; const int M = hparams.n_audio_ctx;
struct ggml_init_params params; struct ggml_init_params params = {
.mem_size = g_buf_compute.size(),
{ .mem_buffer = g_buf_compute.data(),
static size_t buf_size = MEM_REQ_DECODE.at(model.type);
static void * buf = malloc(buf_size);
params = {
.mem_size = buf_size,
.mem_buffer = buf,
}; };
}
struct ggml_context * ctx0 = ggml_init(params); struct ggml_context * ctx0 = ggml_init(params);
@ -1525,17 +1512,10 @@ bool whisper_decode(
for (int il = 0; il < n_layer; ++il) { for (int il = 0; il < n_layer; ++il) {
const auto & layer = model.layers_decoder[il]; const auto & layer = model.layers_decoder[il];
struct ggml_init_params paramsL; struct ggml_init_params paramsL = {
.mem_size = g_buf_compute_layer.size(),
{ .mem_buffer = g_buf_compute_layer.data(),
static size_t buf_size = MEM_REQ_DECODE_LAYER.at(model.type); };
static void * buf = malloc(buf_size);
paramsL = {
.mem_size = buf_size,
.mem_buffer = buf,
};
}
struct ggml_context * ctxL = ggml_init(paramsL); struct ggml_context * ctxL = ggml_init(paramsL);
struct ggml_cgraph gf = { .n_threads = n_threads }; struct ggml_cgraph gf = { .n_threads = n_threads };
@ -1849,7 +1829,7 @@ bool whisper_decode(
// TODO: temperature // TODO: temperature
whisper_vocab::id whisper_sample_best( whisper_vocab::id whisper_sample_best(
const whisper_vocab & vocab, const whisper_vocab & vocab,
const float * probs) { const float * probs, bool need_timestamp) {
int n_logits = vocab.id_to_token.size(); int n_logits = vocab.id_to_token.size();
std::vector<std::pair<double, whisper_vocab::id>> probs_id; std::vector<std::pair<double, whisper_vocab::id>> probs_id;
@ -1859,7 +1839,7 @@ whisper_vocab::id whisper_sample_best(
probs_id.push_back(std::make_pair(probs[i], i)); probs_id.push_back(std::make_pair(probs[i], i));
} }
const int top_k = 10; const int top_k = 4;
// find the top K tokens // find the top K tokens
std::partial_sort( std::partial_sort(
@ -1876,6 +1856,15 @@ whisper_vocab::id whisper_sample_best(
// printf("%d: '%s' %f, %d\n", i, vocab.id_to_token.at(probs_id[i].second).c_str(), probs_id[i].first, probs_id[i].second); // printf("%d: '%s' %f, %d\n", i, vocab.id_to_token.at(probs_id[i].second).c_str(), probs_id[i].first, probs_id[i].second);
//} //}
if (need_timestamp) {
// at the end of the 30-second audio segment, we start giving preference to time tokens
for (int i = 0; i < top_k; i++) {
if (probs_id[i].second > vocab.token_beg + 1300 && probs_id[i].first > probs_id[0].first*0.1) {
return probs_id[i].second;
}
}
}
int res = 0; int res = 0;
while ((probs_id[res].second == vocab.token_sot || while ((probs_id[res].second == vocab.token_sot ||
probs_id[res].second == vocab.token_solm || probs_id[res].second == vocab.token_solm ||
@ -2136,8 +2125,8 @@ int main(int argc, char ** argv) {
return 2; return 2;
} }
if (wav.channels != 1) { if (wav.channels != 1 && wav.channels != 2) {
fprintf(stderr, "%s: WAV file '%s' must be mono\n", argv[0], params.fname_inp.c_str()); fprintf(stderr, "%s: WAV file '%s' must be mono or stereo\n", argv[0], params.fname_inp.c_str());
return 3; return 3;
} }
@ -2158,8 +2147,14 @@ int main(int argc, char ** argv) {
// convert to float // convert to float
pcmf32.resize(pcm16.size()); pcmf32.resize(pcm16.size());
for (size_t i = 0; i < pcm16.size(); i++) { if (wav.channels == 1) {
pcmf32[i] = float(pcm16[i])/32768.0f; for (size_t i = 0; i < pcm16.size(); i++) {
pcmf32[i] = float(pcm16[i])/32768.0f;
}
} else {
for (size_t i = 0; i < pcm16.size(); i++) {
pcmf32[i] = float(pcm16[i*2 + 0] + pcm16[i*2 + 1])/32768.0f/2.0f;
}
} }
} }
@ -2252,6 +2247,7 @@ int main(int argc, char ** argv) {
int seek_delta = 100*CHUNK_SIZE; int seek_delta = 100*CHUNK_SIZE;
whisper_vocab::id last_id = 0; whisper_vocab::id last_id = 0;
// print the prompt
//printf("\n\n"); //printf("\n\n");
//for (int i = 0; i < prompt.size(); i++) { //for (int i = 0; i < prompt.size(); i++) {
// printf("%s: prompt[%d] = %s\n", __func__, i, vocab.id_to_token[prompt[i]].c_str()); // printf("%s: prompt[%d] = %s\n", __func__, i, vocab.id_to_token[prompt[i]].c_str());
@ -2294,7 +2290,7 @@ int main(int argc, char ** argv) {
{ {
const int64_t t_start_sample_us = ggml_time_us(); const int64_t t_start_sample_us = ggml_time_us();
id = whisper_sample_best(vocab, probs.data() + (probs.size() - n_vocab)); id = whisper_sample_best(vocab, probs.data() + (probs.size() - n_vocab), result_len == 0);
if (i > 0) { if (i > 0) {
tid = whisper_sample_timestamp(vocab, probs.data() + (probs.size() - n_vocab)); tid = whisper_sample_timestamp(vocab, probs.data() + (probs.size() - n_vocab));
} }
@ -2313,6 +2309,8 @@ int main(int argc, char ** argv) {
prompt.push_back(id); prompt.push_back(id);
result_cur.push_back({ id, seek + 2*(tid - vocab.token_beg) }); result_cur.push_back({ id, seek + 2*(tid - vocab.token_beg) });
//printf("%s: %s\n", __func__, vocab.id_to_token[id].c_str());
// end of text token // end of text token
if (id == vocab.token_eot) { if (id == vocab.token_eot) {
break; break;