Spaces:
Configuration error
Configuration error
File size: 4,189 Bytes
b56d19a 6472666 b56d19a b4964a2 b56d19a ba81a8e b56d19a ba81a8e b56d19a ba81a8e b56d19a ba81a8e b56d19a 838b1d7 b56d19a |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 |
!!! warning
This feature not supported on ARM devices only x86_64. I was unable to build [piper-phonemize](https://github.com/rhasspy/piper-phonemize)(my [fork](https://github.com/fedirz/piper-phonemize))
TODO: add a note about automatic downloads
TODO: add a demo
TODO: add a note about tts only running on cpu
TODO: add a note about exploring other models
TODO: add a note about performance
!!! note
Before proceeding, make sure you are familiar with the [OpenAI Text-to-Speech](https://platform.openai.com/docs/guides/text-to-speech) and the relevant [OpenAI API reference](https://platform.openai.com/docs/api-reference/audio/createSpeech)
## Prerequisite
Download the piper voices from [HuggingFace model repository](https://huggingface.co/rhasspy/piper-voices)
```bash
# Download all voices (~15 minutes / 7.7 Gbs)
docker exec -it speaches huggingface-cli download rhasspy/piper-voices
# Download all English voices (~4.5 minutes)
docker exec -it speaches huggingface-cli download rhasspy/piper-voices --include 'en/**/*' 'voices.json'
# Download all qualities of a specific voice (~4 seconds)
docker exec -it speaches huggingface-cli download rhasspy/piper-voices --include 'en/en_US/amy/**/*' 'voices.json'
# Download specific quality of a specific voice (~2 seconds)
docker exec -it speaches huggingface-cli download rhasspy/piper-voices --include 'en/en_US/amy/medium/*' 'voices.json'
```
!!! note
You can find audio samples of all the available voices [here](https://rhasspy.github.io/piper-samples/)
## Curl
```bash
# Generate speech from text using the default values (response_format="mp3", speed=1.0, voice="en_US-amy-medium", etc.)
curl http://localhost:8000/v1/audio/speech --header "Content-Type: application/json" --data '{"input": "Hello World!"}' --output audio.mp3
# Specifying the output format
curl http://localhost:8000/v1/audio/speech --header "Content-Type: application/json" --data '{"input": "Hello World!", "response_format": "wav"}' --output audio.wav
# Specifying the audio speed
curl http://localhost:8000/v1/audio/speech --header "Content-Type: application/json" --data '{"input": "Hello World!", "speed": 2.0}' --output audio.mp3
# List available (downloaded) voices
curl http://localhost:8000/v1/audio/speech/voices
# List just the voice names
curl http://localhost:8000/v1/audio/speech/voices | jq --raw-output '.[] | .voice'
# List just the voices in your language
curl --silent http://localhost:8000/v1/audio/speech/voices | jq --raw-output '.[] | select(.voice | startswith("en")) | .voice'
curl http://localhost:8000/v1/audio/speech --header "Content-Type: application/json" --data '{"input": "Hello World!", "voice": "en_US-ryan-high"}' --output audio.mp3
```
## Python
=== "httpx"
```python
from pathlib import Path
import httpx
client = httpx.Client(base_url="http://localhost:8000/")
res = client.post(
"v1/audio/speech",
json={
"model": "piper",
"voice": "en_US-amy-medium",
"input": "Hello, world!",
"response_format": "mp3",
"speed": 1,
},
).raise_for_status()
with Path("output.mp3").open("wb") as f:
f.write(res.read())
```
## OpenAI SDKs
!!! note
Although this project doesn't require an API key, all OpenAI SDKs require an API key. Therefore, you will need to set it to a non-empty value. Additionally, you will need to overwrite the base URL to point to your server.
This can be done by setting the `OPENAI_API_KEY` and `OPENAI_BASE_URL` environment variables or by passing them as arguments to the SDK.
=== "Python"
```python
from pathlib import Path
from openai import OpenAI
openai = OpenAI(base_url="http://localhost:8000/v1", api_key="cant-be-empty")
res = openai.audio.speech.create(
model="piper",
voice="en_US-amy-medium", # pyright: ignore[reportArgumentType]
input="Hello, world!",
response_format="mp3",
speed=1,
)
with Path("output.mp3").open("wb") as f:
f.write(res.response.read())
```
=== "Other"
See [OpenAI libraries](https://platform.openai.com/docs/libraries)
|