speech

package

v0.3.2 Latest Latest Go to latest Published: Nov 3, 2026 License: GPL-3.0 Imports: 13 Imported by: 0

Details

Valid go.mod file
Redistributable license
Tagged version
Stable version
Learn more about best practices

Repository

github.com/braedonsaunders/ether

Links

Open Source Insights

Documentation ¶

Overview ¶

Package speech sends audio to an OpenAI-compatible transcription endpoint (Whisper and the servers that speak its API). Callers invoke it only after a viewer asks for a subtitle track.

Index ¶

func Endpoint(base string) string
func ExtractWAV(ctx context.Context, ffmpegPath, input string) (path string, cleanup func(), err error)
type Client
- func (c Client) Transcribe(ctx context.Context, audio io.Reader, filename, language string) (string, error)
- func (c Client) TranscribeFile(ctx context.Context, wavPath, language string) (string, error)

Constants ¶

This section is empty.

Variables ¶

This section is empty.

Functions ¶

func Endpoint ¶

func Endpoint(base string) string

Endpoint is the transcription URL for an owner-entered base. "http://host:1234" and "http://host:1234/v1" both work.

func ExtractWAV ¶

func ExtractWAV(ctx context.Context, ffmpegPath, input string) (path string, cleanup func(), err error)

ExtractWAV writes 16 kHz mono PCM wav for input (a file or URL ffmpeg can read). cleanup removes the file. The caller asks for this only after a viewer requests subtitles.

Types ¶

type Client ¶

type Client struct {
	BaseURL string
	APIKey  string
	Model   string
	HTTP    *http.Client
}

Client talks to /audio/transcriptions.

func (Client) Transcribe ¶

func (c Client) Transcribe(ctx context.Context, audio io.Reader, filename, language string) (string, error)

Transcribe posts one audio file and returns the response body. response_format=vtt asks for timed cues. filename is the multipart name.

func (Client) TranscribeFile ¶

func (c Client) TranscribeFile(ctx context.Context, wavPath, language string) (string, error)

TranscribeFile sends a wav, splitting it into 10-minute pieces when it is larger than one request should carry, and returns one WebVTT document.

Source Files ¶

View all Source files

?	: This menu
/	: Search site
f or F	: Jump to
y or Y	: Canonical URL